diff --git a/.gitignore b/.gitignore
index 78d8da20e478..85ca98724dc2 100644
--- a/.gitignore
+++ b/.gitignore
@@ -55,6 +55,8 @@ tensorrt_llm/scripts
 *docs/source/_cpp_gen*
 docs/source/**/*.rst
 !docs/source/examples/index.rst
+!docs/source/deployment-guide/config_table.rst
+!docs/source/deployment-guide/note_sections.rst
 *.swp
 
 # Testing
diff --git a/docs/source/deployment-guide/config_table.rst b/docs/source/deployment-guide/config_table.rst
new file mode 100644
index 000000000000..d28fed25a8ea
--- /dev/null
+++ b/docs/source/deployment-guide/config_table.rst
@@ -0,0 +1,1074 @@
+.. include:: note_sections.rst
+   :start-after: .. start-note-traffic-patterns
+   :end-before: .. end-note-traffic-patterns
+
+.. start-deepseek-ai/DeepSeek-R1-0528
+
+.. _deepseek-ai/DeepSeek-R1-0528:
+
+`DeepSeek-R1 <https://huggingface.co/deepseek-ai/DeepSeek-R1-0528>`_
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. list-table::
+   :width: 100%
+   :header-rows: 1
+   :widths: 12 15 15 13 20 25
+
+   * - GPU
+     - Performance Profile
+     - ISL / OSL
+     - Concurrency
+     - Config
+     - Command
+   * - 8xB200_NVL
+     - Min Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc8.yaml``
+   * - 8xB200_NVL
+     - Balanced
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc16.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc32.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc64.yaml``
+   * - 8xB200_NVL
+     - Min Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc8.yaml``
+   * - 8xB200_NVL
+     - Balanced
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc16.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc32.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc64.yaml``
+   * - 8xH200_SXM
+     - Min Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc4.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc8.yaml``
+   * - 8xH200_SXM
+     - Balanced
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc16.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc32.yaml``
+   * - 8xH200_SXM
+     - Max Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc64.yaml``
+   * - 8xH200_SXM
+     - Min Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc4.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc8.yaml``
+   * - 8xH200_SXM
+     - Balanced
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc16.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc32.yaml``
+   * - 8xH200_SXM
+     - Max Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc64.yaml``
+
+.. end-deepseek-ai/DeepSeek-R1-0528
+
+.. start-nvidia/DeepSeek-R1-0528-FP4-v2
+
+.. _nvidia/DeepSeek-R1-0528-FP4-v2:
+
+`DeepSeek-R1 (NVFP4) <https://huggingface.co/nvidia/DeepSeek-R1-0528-FP4-v2>`_
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. list-table::
+   :width: 100%
+   :header-rows: 1
+   :widths: 12 15 15 13 20 25
+
+   * - GPU
+     - Performance Profile
+     - ISL / OSL
+     - Concurrency
+     - Config
+     - Command
+   * - 4xB200_NVL
+     - Min Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc4.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc4.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc8.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc8.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc8.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc16.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc16.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc16.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc32.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc32.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc32.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc64.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc64.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc64.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 128
+     - `1k1k_tp4_conc128.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc128.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc128.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 128
+     - `1k1k_tp8_conc128.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc128.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc128.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 256
+     - `1k1k_tp4_conc256.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc256.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc256.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 1024 / 1024
+     - 256
+     - `1k1k_tp8_conc256.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc256.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc256.yaml``
+   * - 4xB200_NVL
+     - Min Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc4.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc4.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc8.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc8.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc8.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc16.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc16.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc16.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc32.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc32.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc32.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc64.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc64.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc64.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 128
+     - `8k1k_tp4_conc128.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc128.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc128.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 128
+     - `8k1k_tp8_conc128.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc128.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc128.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 256
+     - `8k1k_tp4_conc256.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc256.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc256.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 8192 / 1024
+     - 256
+     - `8k1k_tp8_conc256.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc256.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-0528-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc256.yaml``
+
+.. end-nvidia/DeepSeek-R1-0528-FP4-v2
+
+.. start-openai/gpt-oss-120b
+
+.. _openai/gpt-oss-120b:
+
+`gpt-oss-120b <https://huggingface.co/openai/gpt-oss-120b>`_
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. list-table::
+   :width: 100%
+   :header-rows: 1
+   :widths: 12 15 15 13 20 25
+
+   * - GPU
+     - Performance Profile
+     - ISL / OSL
+     - Concurrency
+     - Config
+     - Command
+   * - B200_NVL
+     - Min Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp1_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc4.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp2_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc4.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc4.yaml``
+   * - B200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp1_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc8.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp2_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc8.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc8.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc8.yaml``
+   * - B200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp1_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc16.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp2_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc16.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc16.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc16.yaml``
+   * - B200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp1_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc32.yaml``
+   * - 2xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp2_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc32.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc32.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc32.yaml``
+   * - B200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp1_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc64.yaml``
+   * - 2xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp2_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc64.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc64.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc64.yaml``
+   * - B200_NVL
+     - Min Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp1_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc4.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp2_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc4.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc4.yaml``
+   * - B200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp1_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc8.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp2_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc8.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc8.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc8.yaml``
+   * - B200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp1_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc16.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp2_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc16.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc16.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc16.yaml``
+   * - B200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp1_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc32.yaml``
+   * - 2xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp2_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc32.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc32.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc32.yaml``
+   * - B200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp1_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc64.yaml``
+   * - 2xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp2_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc64.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc64.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc64.yaml``
+   * - B200_NVL
+     - Min Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp1_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc4.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp2_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc4.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc4.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc4.yaml``
+   * - B200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp1_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc8.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp2_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc8.yaml``
+   * - 4xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc8.yaml``
+   * - 8xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc8.yaml``
+   * - B200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp1_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc16.yaml``
+   * - 2xB200_NVL
+     - Low Latency
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp2_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc16.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc16.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc16.yaml``
+   * - B200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp1_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc32.yaml``
+   * - 2xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp2_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc32.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc32.yaml``
+   * - 8xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc32.yaml``
+   * - B200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp1_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc64.yaml``
+   * - 2xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp2_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc64.yaml``
+   * - 4xB200_NVL
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc64.yaml``
+   * - 8xB200_NVL
+     - Max Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc64.yaml``
+   * - H200_SXM
+     - Min Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp1_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc4.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp2_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc4.yaml``
+   * - 4xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc4.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 4
+     - `1k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc4.yaml``
+   * - H200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp1_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc8.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp2_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc8.yaml``
+   * - 4xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc8.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 8
+     - `1k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc8.yaml``
+   * - H200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp1_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc16.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp2_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc16.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc16.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 16
+     - `1k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc16.yaml``
+   * - H200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp1_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc32.yaml``
+   * - 2xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp2_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc32.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc32.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 32
+     - `1k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc32.yaml``
+   * - H200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp1_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc64.yaml``
+   * - 2xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp2_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc64.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc64.yaml``
+   * - 8xH200_SXM
+     - Max Throughput
+     - 1024 / 1024
+     - 64
+     - `1k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc64.yaml``
+   * - H200_SXM
+     - Min Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp1_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc4.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp2_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc4.yaml``
+   * - 4xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc4.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 4
+     - `1k8k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc4.yaml``
+   * - H200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp1_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc8.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp2_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc8.yaml``
+   * - 4xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc8.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 8
+     - `1k8k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc8.yaml``
+   * - H200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp1_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc16.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp2_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc16.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc16.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 16
+     - `1k8k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc16.yaml``
+   * - H200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp1_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc32.yaml``
+   * - 2xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp2_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc32.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc32.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 32
+     - `1k8k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc32.yaml``
+   * - H200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp1_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc64.yaml``
+   * - 2xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp2_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc64.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc64.yaml``
+   * - 8xH200_SXM
+     - Max Throughput
+     - 1024 / 8192
+     - 64
+     - `1k8k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc64.yaml``
+   * - H200_SXM
+     - Min Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp1_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc4.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp2_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc4.yaml``
+   * - 4xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp4_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc4.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 4
+     - `8k1k_tp8_conc4.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc4.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc4.yaml``
+   * - H200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp1_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc8.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp2_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc8.yaml``
+   * - 4xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp4_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc8.yaml``
+   * - 8xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 8
+     - `8k1k_tp8_conc8.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc8.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc8.yaml``
+   * - H200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp1_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc16.yaml``
+   * - 2xH200_SXM
+     - Low Latency
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp2_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc16.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp4_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc16.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 16
+     - `8k1k_tp8_conc16.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc16.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc16.yaml``
+   * - H200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp1_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc32.yaml``
+   * - 2xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp2_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc32.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp4_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc32.yaml``
+   * - 8xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 32
+     - `8k1k_tp8_conc32.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc32.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc32.yaml``
+   * - H200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp1_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc64.yaml``
+   * - 2xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp2_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc64.yaml``
+   * - 4xH200_SXM
+     - High Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp4_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc64.yaml``
+   * - 8xH200_SXM
+     - Max Throughput
+     - 8192 / 1024
+     - 64
+     - `8k1k_tp8_conc64.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc64.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc64.yaml``
+
+.. end-openai/gpt-oss-120b
diff --git a/docs/source/deployment-guide/deployment-guide-for-deepseek-r1-on-trtllm.md b/docs/source/deployment-guide/deployment-guide-for-deepseek-r1-on-trtllm.md
index 3a25c5c7526e..e4165eac09c6 100644
--- a/docs/source/deployment-guide/deployment-guide-for-deepseek-r1-on-trtllm.md
+++ b/docs/source/deployment-guide/deployment-guide-for-deepseek-r1-on-trtllm.md
@@ -66,7 +66,7 @@ We maintain YAML configuration files with recommended performance settings in th
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/deepseek-r1-throughput.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/deepseek-r1-throughput.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -74,7 +74,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/deepseek-r1-throughput.yaml
+```{literalinclude} ../../../examples/configs/curated/deepseek-r1-throughput.yaml
 ---
 language: shell
 prepend: |
@@ -90,7 +90,7 @@ To use the `DeepGEMM` MOE backend on B200/GB200, use this config instead:
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/deepseek-r1-deepgemm.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/deepseek-r1-deepgemm.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -98,7 +98,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/deepseek-r1-deepgemm.yaml
+```{literalinclude} ../../../examples/configs/curated/deepseek-r1-deepgemm.yaml
 ---
 language: shell
 prepend: |
@@ -154,7 +154,7 @@ These options provide control over TensorRT LLM's behavior and are set within th
 
 #### `trust_remote_code`
 
-&emsp;**Description:** Allows TensorRT LLM to download models and tokenizers from Hugging Face. This flag is passed directly to the Hugging Face API.
+* **Description:** Allows TensorRT LLM to download models and tokenizers from Hugging Face. This flag is passed directly to the Hugging Face API.
 
 #### `kv_cache_config`
 
@@ -429,3 +429,23 @@ $$
 $$
 \text{TPS} = \frac{\text{Num Output Tokens}}{T_{last} - T_{first}}
 $$
+
+## Preconfigured Recipes
+
+The following tables list recommended configurations from the comprehensive database for different performance profiles.
+
+```{eval-rst}
+.. include:: note_sections.rst
+   :start-after: .. start-note-traffic-patterns
+   :end-before: .. end-note-traffic-patterns
+
+.. include:: config_table.rst
+   :start-after: .. start-deepseek-ai/DeepSeek-R1-0528
+   :end-before: .. end-deepseek-ai/DeepSeek-R1-0528
+```
+
+```{eval-rst}
+.. include:: config_table.rst
+   :start-after: .. start-nvidia/DeepSeek-R1-0528-FP4-v2
+   :end-before: .. end-nvidia/DeepSeek-R1-0528-FP4-v2
+```
diff --git a/docs/source/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md b/docs/source/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md
index 86fc4bc786ad..5a9f9f4c7263 100644
--- a/docs/source/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md
+++ b/docs/source/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md
@@ -64,7 +64,7 @@ For low-latency use cases:
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/gpt-oss-120b-latency.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/gpt-oss-120b-latency.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -72,7 +72,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/gpt-oss-120b-latency.yaml
+```{literalinclude} ../../../examples/configs/curated/gpt-oss-120b-latency.yaml
 ---
 language: shell
 prepend: |
@@ -88,7 +88,7 @@ For max-throughput use cases:
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/gpt-oss-120b-throughput.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/gpt-oss-120b-throughput.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -96,7 +96,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/gpt-oss-120b-throughput.yaml
+```{literalinclude} ../../../examples/configs/curated/gpt-oss-120b-throughput.yaml
 ---
 language: shell
 prepend: |
@@ -377,3 +377,17 @@ $$
 $$
 \text{TPS} = \frac{\text{Num Output Tokens}}{T_{last} - T_{first}}
 $$
+
+## Preconfigured Recipes
+
+The following table lists recommended configurations from the comprehensive database for different performance profiles.
+
+```{eval-rst}
+.. include:: note_sections.rst
+   :start-after: .. start-note-traffic-patterns
+   :end-before: .. end-note-traffic-patterns
+
+.. include:: config_table.rst
+   :start-after: .. start-openai/gpt-oss-120b
+   :end-before: .. end-openai/gpt-oss-120b
+```
diff --git a/docs/source/deployment-guide/deployment-guide-for-llama3.3-70b-on-trtllm.md b/docs/source/deployment-guide/deployment-guide-for-llama3.3-70b-on-trtllm.md
index 583ef56b4939..d3e328d810dc 100644
--- a/docs/source/deployment-guide/deployment-guide-for-llama3.3-70b-on-trtllm.md
+++ b/docs/source/deployment-guide/deployment-guide-for-llama3.3-70b-on-trtllm.md
@@ -58,7 +58,7 @@ We maintain YAML configuration files with recommended performance settings in th
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/llama-3.3-70b.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/llama-3.3-70b.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -66,7 +66,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/llama-3.3-70b.yaml
+```{literalinclude} ../../../examples/configs/curated/llama-3.3-70b.yaml
 ---
 language: shell
 prepend: |
diff --git a/docs/source/deployment-guide/deployment-guide-for-llama4-scout-on-trtllm.md b/docs/source/deployment-guide/deployment-guide-for-llama4-scout-on-trtllm.md
index 10db2e128f39..7d69b7a8be79 100644
--- a/docs/source/deployment-guide/deployment-guide-for-llama4-scout-on-trtllm.md
+++ b/docs/source/deployment-guide/deployment-guide-for-llama4-scout-on-trtllm.md
@@ -57,7 +57,7 @@ We maintain YAML configuration files with recommended performance settings in th
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/llama-4-scout.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/llama-4-scout.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -65,7 +65,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/llama-4-scout.yaml
+```{literalinclude} ../../../examples/configs/curated/llama-4-scout.yaml
 ---
 language: shell
 prepend: |
diff --git a/docs/source/deployment-guide/deployment-guide-for-qwen3-next-on-trtllm.md b/docs/source/deployment-guide/deployment-guide-for-qwen3-next-on-trtllm.md
index 246fc74a567c..46bf724b7131 100644
--- a/docs/source/deployment-guide/deployment-guide-for-qwen3-next-on-trtllm.md
+++ b/docs/source/deployment-guide/deployment-guide-for-qwen3-next-on-trtllm.md
@@ -35,7 +35,7 @@ We maintain YAML configuration files with recommended performance settings in th
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/qwen3-next.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/qwen3-next.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -43,7 +43,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/qwen3-next.yaml
+```{literalinclude} ../../../examples/configs/curated/qwen3-next.yaml
 ---
 language: shell
 prepend: |
diff --git a/docs/source/deployment-guide/deployment-guide-for-qwen3-on-trtllm.md b/docs/source/deployment-guide/deployment-guide-for-qwen3-on-trtllm.md
index 190740ebd858..894c6a1e63f1 100644
--- a/docs/source/deployment-guide/deployment-guide-for-qwen3-on-trtllm.md
+++ b/docs/source/deployment-guide/deployment-guide-for-qwen3-on-trtllm.md
@@ -40,7 +40,7 @@ We maintain YAML configuration files with recommended performance settings in th
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/qwen3.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/qwen3.yaml
 ```
 
 Note: if you don't have access to the source code locally, you can manually create the YAML config file using the code in the dropdown below.
@@ -48,7 +48,7 @@ Note: if you don't have access to the source code locally, you can manually crea
 ````{admonition} Show code
 :class: dropdown
 
-```{literalinclude} ../../../examples/configs/qwen3.yaml
+```{literalinclude} ../../../examples/configs/curated/qwen3.yaml
 ---
 language: shell
 prepend: |
diff --git a/docs/source/deployment-guide/index.rst b/docs/source/deployment-guide/index.rst
index ed7fd9c536ef..644a9d9ae95e 100644
--- a/docs/source/deployment-guide/index.rst
+++ b/docs/source/deployment-guide/index.rst
@@ -6,15 +6,20 @@ Quick Start for Popular Models
 
 The table below contains ``trtllm-serve`` commands that can be used to easily deploy popular models including DeepSeek-R1, gpt-oss, Llama 4, Qwen3, and more.
 
-We maintain LLM API configuration files for these models containing recommended performance settings in the `examples/configs <https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/configs>`_ directory. The TensorRT LLM Docker container makes the config files available at ``/app/tensorrt_llm/examples/configs``, but you can customize this as needed:
+We maintain LLM API configuration files for these models containing recommended performance settings in two locations:
+
+* **Curated Examples**: `examples/configs/curated <https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/configs/curated>`_ - Hand-picked configurations for common scenarios.
+* **Comprehensive Database**: `examples/configs/database <https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/configs/database>`_ - A more comprehensive set of known-good configurations for various GPUs and traffic patterns.
+
+The TensorRT LLM Docker container makes these config files available at ``/app/tensorrt_llm/examples/configs/curated`` and ``/app/tensorrt_llm/examples/configs/database`` respectively. You can reference them as needed:
 
 .. code-block:: bash
 
    export TRTLLM_DIR="/app/tensorrt_llm" # path to the TensorRT LLM repo in your local environment
 
-.. note::
-
-   The configs here are specifically optimized for a target ISL/OSL (Input/Output Sequence Length) of 1024/1024. If your traffic pattern is different, you may benefit from additional tuning. In the future, we plan to provide more configs for a wider range of traffic patterns.
+.. include:: note_sections.rst
+   :start-after: .. start-note-quick-start-isl-osl
+   :end-before: .. end-note-quick-start-isl-osl
 
 This table is designed to provide a straightforward starting point; for detailed model-specific deployment guides, check out the guides below.
 
@@ -30,53 +35,53 @@ This table is designed to provide a straightforward starting point; for detailed
    * - `DeepSeek-R1 <https://huggingface.co/deepseek-ai/DeepSeek-R1-0528>`_
      - H100, H200
      - Max Throughput
-     - `deepseek-r1-throughput.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/deepseek-r1-throughput.yaml>`_
-     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/deepseek-r1-throughput.yaml``
+     - `deepseek-r1-throughput.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/deepseek-r1-throughput.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/deepseek-r1-throughput.yaml``
    * - `DeepSeek-R1 <https://huggingface.co/deepseek-ai/DeepSeek-R1-0528>`_
      - B200, GB200
      - Max Throughput
-     - `deepseek-r1-deepgemm.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/deepseek-r1-deepgemm.yaml>`_
-     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/deepseek-r1-deepgemm.yaml``
+     - `deepseek-r1-deepgemm.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/deepseek-r1-deepgemm.yaml>`_
+     - ``trtllm-serve deepseek-ai/DeepSeek-R1-0528 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/deepseek-r1-deepgemm.yaml``
    * - `DeepSeek-R1 (NVFP4) <https://huggingface.co/nvidia/DeepSeek-R1-FP4>`_
      - B200, GB200
      - Max Throughput
-     - `deepseek-r1-throughput.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/deepseek-r1-throughput.yaml>`_
-     - ``trtllm-serve nvidia/DeepSeek-R1-FP4 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/deepseek-r1-throughput.yaml``
+     - `deepseek-r1-throughput.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/deepseek-r1-throughput.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-FP4 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/deepseek-r1-throughput.yaml``
    * - `DeepSeek-R1 (NVFP4) <https://huggingface.co/nvidia/DeepSeek-R1-FP4-v2>`_
      - B200, GB200
      - Min Latency
-     - `deepseek-r1-latency.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/deepseek-r1-latency.yaml>`_
-     - ``trtllm-serve nvidia/DeepSeek-R1-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/deepseek-r1-latency.yaml``
+     - `deepseek-r1-latency.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/deepseek-r1-latency.yaml>`_
+     - ``trtllm-serve nvidia/DeepSeek-R1-FP4-v2 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/deepseek-r1-latency.yaml``
    * - `gpt-oss-120b <https://huggingface.co/openai/gpt-oss-120b>`_
      - Any
      - Max Throughput
-     - `gpt-oss-120b-throughput.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/gpt-oss-120b-throughput.yaml>`_
-     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/gpt-oss-120b-throughput.yaml``
+     - `gpt-oss-120b-throughput.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/gpt-oss-120b-throughput.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/gpt-oss-120b-throughput.yaml``
    * - `gpt-oss-120b <https://huggingface.co/openai/gpt-oss-120b>`_
      - Any
      - Min Latency
-     - `gpt-oss-120b-latency.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/gpt-oss-120b-latency.yaml>`_
-     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/gpt-oss-120b-latency.yaml``
+     - `gpt-oss-120b-latency.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/gpt-oss-120b-latency.yaml>`_
+     - ``trtllm-serve openai/gpt-oss-120b --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/gpt-oss-120b-latency.yaml``
    * - `Qwen3-Next-80B-A3B-Thinking <https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking>`_
      - Any
      - Max Throughput
-     - `qwen3-next.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/qwen3-next.yaml>`_
-     - ``trtllm-serve Qwen/Qwen3-Next-80B-A3B-Thinking --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/qwen3-next.yaml``
+     - `qwen3-next.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/qwen3-next.yaml>`_
+     - ``trtllm-serve Qwen/Qwen3-Next-80B-A3B-Thinking --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/qwen3-next.yaml``
    * - Qwen3 family (e.g. `Qwen3-30B-A3B <https://huggingface.co/Qwen/Qwen3-30B-A3B>`_)
      - Any
      - Max Throughput
-     - `qwen3.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/qwen3.yaml>`_
-     - ``trtllm-serve Qwen/Qwen3-30B-A3B --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/qwen3.yaml`` (swap to another Qwen3 model name as needed)
+     - `qwen3.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/qwen3.yaml>`_
+     - ``trtllm-serve Qwen/Qwen3-30B-A3B --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/qwen3.yaml`` (swap to another Qwen3 model name as needed)
    * - `Llama-3.3-70B (FP8) <https://huggingface.co/nvidia/Llama-3.3-70B-Instruct-FP8>`_
      - Any
      - Max Throughput
-     - `llama-3.3-70b.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/llama-3.3-70b.yaml>`_
-     - ``trtllm-serve nvidia/Llama-3.3-70B-Instruct-FP8 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/llama-3.3-70b.yaml``
+     - `llama-3.3-70b.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/llama-3.3-70b.yaml>`_
+     - ``trtllm-serve nvidia/Llama-3.3-70B-Instruct-FP8 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/llama-3.3-70b.yaml``
    * - `Llama 4 Scout (FP8) <https://huggingface.co/nvidia/Llama-4-Scout-17B-16E-Instruct-FP8>`_
      - Any
      - Max Throughput
-     - `llama-4-scout.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/llama-4-scout.yaml>`_
-     - ``trtllm-serve nvidia/Llama-4-Scout-17B-16E-Instruct-FP8 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/llama-4-scout.yaml``
+     - `llama-4-scout.yaml <https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/configs/curated/llama-4-scout.yaml>`_
+     - ``trtllm-serve nvidia/Llama-4-Scout-17B-16E-Instruct-FP8 --extra_llm_api_options ${TRTLLM_DIR}/examples/configs/curated/llama-4-scout.yaml``
 
 Model-Specific Deployment Guides
 ---------------------------------
@@ -94,3 +99,10 @@ The deployment guides below provide more detailed instructions for serving speci
    deployment-guide-for-qwen3-on-trtllm.md
    deployment-guide-for-qwen3-next-on-trtllm.md
    deployment-guide-for-kimi-k2-thinking-on-trtllm.md
+
+Comprehensive Configuration Database
+------------------------------------
+
+The table below lists all available pre-configured model scenarios in the TensorRT LLM configuration database. Each row represents a specific model, GPU, and performance profile combination with recommended request settings.
+
+.. include:: config_table.rst
diff --git a/docs/source/deployment-guide/note_sections.rst b/docs/source/deployment-guide/note_sections.rst
new file mode 100644
index 000000000000..4cd0d1c41dda
--- /dev/null
+++ b/docs/source/deployment-guide/note_sections.rst
@@ -0,0 +1,36 @@
+..
+   Reusable note sections for deployment guides.
+   Include specific notes using:
+
+   .. include:: note_sections.rst
+      :start-after: .. start-note-<name>
+      :end-before: .. end-note-<name>
+
+.. start-note-traffic-patterns
+
+.. note::
+
+   **Traffic Patterns**: The ISL (Input Sequence Length) and OSL (Output Sequence Length)
+   values in each configuration represent the **maximum supported values** for that config.
+   Requests exceeding these limits may result in errors.
+
+   To handle requests with input sequences **longer than the configured ISL**, add the following
+   to your config file:
+
+   .. code-block:: yaml
+
+      enable_chunked_prefill: true
+
+   This enables chunked prefill, which processes long input sequences in chunks rather than
+   requiring them to fit within a single prefill operation. Note that enabling chunked prefill
+   does **not** guarantee optimal performance—these configs are tuned for the specified ISL/OSL.
+
+.. end-note-traffic-patterns
+
+.. start-note-quick-start-isl-osl
+
+.. note::
+
+   The configs here are specifically optimized for a target ISL/OSL (Input/Output Sequence Length) of 1024/1024. If your traffic pattern is different, refer to the :ref:`Comprehensive Configuration Database` section below which covers a larger set of traffic patterns and performance profiles.
+
+.. end-note-quick-start-isl-osl
diff --git a/examples/configs/deepseek-r1-deepgemm.yaml b/examples/configs/curated/deepseek-r1-deepgemm.yaml
similarity index 100%
rename from examples/configs/deepseek-r1-deepgemm.yaml
rename to examples/configs/curated/deepseek-r1-deepgemm.yaml
diff --git a/examples/configs/deepseek-r1-latency.yaml b/examples/configs/curated/deepseek-r1-latency.yaml
similarity index 100%
rename from examples/configs/deepseek-r1-latency.yaml
rename to examples/configs/curated/deepseek-r1-latency.yaml
diff --git a/examples/configs/deepseek-r1-throughput.yaml b/examples/configs/curated/deepseek-r1-throughput.yaml
similarity index 100%
rename from examples/configs/deepseek-r1-throughput.yaml
rename to examples/configs/curated/deepseek-r1-throughput.yaml
diff --git a/examples/configs/gpt-oss-120b-latency.yaml b/examples/configs/curated/gpt-oss-120b-latency.yaml
similarity index 100%
rename from examples/configs/gpt-oss-120b-latency.yaml
rename to examples/configs/curated/gpt-oss-120b-latency.yaml
diff --git a/examples/configs/gpt-oss-120b-throughput.yaml b/examples/configs/curated/gpt-oss-120b-throughput.yaml
similarity index 100%
rename from examples/configs/gpt-oss-120b-throughput.yaml
rename to examples/configs/curated/gpt-oss-120b-throughput.yaml
diff --git a/examples/configs/llama-3.3-70b.yaml b/examples/configs/curated/llama-3.3-70b.yaml
similarity index 100%
rename from examples/configs/llama-3.3-70b.yaml
rename to examples/configs/curated/llama-3.3-70b.yaml
diff --git a/examples/configs/llama-4-scout.yaml b/examples/configs/curated/llama-4-scout.yaml
similarity index 100%
rename from examples/configs/llama-4-scout.yaml
rename to examples/configs/curated/llama-4-scout.yaml
diff --git a/examples/configs/qwen3-disagg-prefill.yaml b/examples/configs/curated/qwen3-disagg-prefill.yaml
similarity index 100%
rename from examples/configs/qwen3-disagg-prefill.yaml
rename to examples/configs/curated/qwen3-disagg-prefill.yaml
diff --git a/examples/configs/qwen3-next.yaml b/examples/configs/curated/qwen3-next.yaml
similarity index 100%
rename from examples/configs/qwen3-next.yaml
rename to examples/configs/curated/qwen3-next.yaml
diff --git a/examples/configs/qwen3.yaml b/examples/configs/curated/qwen3.yaml
similarity index 100%
rename from examples/configs/qwen3.yaml
rename to examples/configs/curated/qwen3.yaml
diff --git a/examples/configs/database/database.py b/examples/configs/database/database.py
new file mode 100644
index 000000000000..e0c73a8ef1ca
--- /dev/null
+++ b/examples/configs/database/database.py
@@ -0,0 +1,64 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+from pathlib import Path
+from typing import Any, Dict, Iterator, List
+
+import yaml
+from pydantic import BaseModel, Field, RootModel
+
+DATABASE_LIST_PATH = Path(__file__).parent / "lookup.yaml"
+
+
+class RecipeConstraints(BaseModel):
+    """Recipe record for scenario list."""
+
+    model: str = Field(description="Model name")
+    gpu: str = Field(description="GPU name")
+    isl: int = Field(description="Input sequence length")
+    osl: int = Field(description="Output sequence length")
+    concurrency: int = Field(description="Concurrency")
+    config_path: str = Field(description="Configuration path")
+    num_gpus: int = Field(description="Number of GPUs")
+
+    def load_config(self) -> Dict[str, Any]:
+        """Load and return the YAML config at config_path."""
+        with open(self.config_path) as f:
+            data = yaml.safe_load(f)
+        return data if data is not None else {}
+
+
+class Recipe(BaseModel):
+    """Recipe that describes a single scenario."""
+
+    constraints: RecipeConstraints = Field(description="Recipe constraints")
+    env_overrides: Dict[str, Any] = Field(description="Environment overrides", default_factory=dict)
+    config: Dict[str, Any] = Field(description="Configuration overrides", default_factory=dict)
+
+
+class RecipeList(RootModel[List[RecipeConstraints]]):
+    @classmethod
+    def from_yaml(cls, yaml_path: Path) -> "RecipeList":
+        """Load and validate recipe list from YAML file."""
+        with open(yaml_path) as f:
+            data = yaml.safe_load(f)
+        return cls(data)
+
+    def __iter__(self) -> Iterator[RecipeConstraints]:
+        return iter(self.root)
+
+    def __len__(self) -> int:
+        return len(self.root)
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc16.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..f770a6566e93
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc32.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..f770a6566e93
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc4.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..f770a6566e93
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc64.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..f770a6566e93
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc64.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc8.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..f770a6566e93
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc16.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..6660bcea96ae
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc32.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..6660bcea96ae
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc4.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..6660bcea96ae
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc64.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..919a0284099e
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc64.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc8.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..6660bcea96ae
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 256
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: DEEPGEMM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc16.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..008da1df54c4
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc32.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..008da1df54c4
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc4.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..008da1df54c4
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc64.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..008da1df54c4
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc64.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc8.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..008da1df54c4
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc16.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..decbb1744a3a
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc32.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..decbb1744a3a
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc4.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..decbb1744a3a
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc64.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..363eebf52142
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc64.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc8.yaml b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..decbb1744a3a
--- /dev/null
+++ b/examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 128
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.75
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/lookup.yaml b/examples/configs/database/lookup.yaml
new file mode 100644
index 000000000000..d1ac7143ceb9
--- /dev/null
+++ b/examples/configs/database/lookup.yaml
@@ -0,0 +1,1176 @@
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 128
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc128.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc16.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 256
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc256.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc32.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc4.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc64.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc8.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 128
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc128.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 256
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc256.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 128
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc128.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc16.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 256
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc256.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc32.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc4.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc64.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc8.yaml
+  num_gpus: 4
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 128
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc128.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 256
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc256.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: nvidia/DeepSeek-R1-0528-FP4-v2
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/1k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/B200/8k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/1k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: deepseek-ai/DeepSeek-R1-0528
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/deepseek-ai/DeepSeek-R1-0528/H200/8k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc16.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc32.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc4.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc64.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc8.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc16.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc32.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc4.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc64.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc8.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc16.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc32.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc4.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc64.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc8.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc16.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc32.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc4.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc64.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc8.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc16.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc32.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc4.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc64.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc8.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc16.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc32.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc4.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc64.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc8.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc16.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc32.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc4.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc64.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc8.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc16.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc32.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc4.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc64.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc8.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc16.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc32.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc4.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc64.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc8.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc16.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc32.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc4.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc64.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc8.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: B200_NVL
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc16.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc32.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc4.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc64.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc8.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc16.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc32.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc4.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc64.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc8.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc16.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc32.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc4.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc64.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc8.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc8.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc16.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc32.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc4.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc64.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc8.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc16.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc32.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc4.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc64.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc8.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc16.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc32.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc4.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc64.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc8.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc16.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc32.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc4.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc64.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 1024
+  osl: 8192
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc8.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc16.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc32.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc4.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc64.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc8.yaml
+  num_gpus: 1
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc16.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc32.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc4.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc64.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc8.yaml
+  num_gpus: 2
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc16.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc32.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc4.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc64.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc8.yaml
+  num_gpus: 4
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 16
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc16.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 32
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc32.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 4
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc4.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 64
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc64.yaml
+  num_gpus: 8
+- model: openai/gpt-oss-120b
+  gpu: H200_SXM
+  isl: 8192
+  osl: 1024
+  concurrency: 8
+  config_path: examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc8.yaml
+  num_gpus: 8
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc128.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc128.yaml
new file mode 100644
index 000000000000..c61e3abc15ed
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc128.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1216
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc16.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc16.yaml
new file mode 100644
index 000000000000..fe58a6a32b27
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc256.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc256.yaml
new file mode 100644
index 000000000000..2a06d3978d01
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc256.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1344
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc32.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc32.yaml
new file mode 100644
index 000000000000..fe58a6a32b27
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc4.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc4.yaml
new file mode 100644
index 000000000000..fe58a6a32b27
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc64.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc64.yaml
new file mode 100644
index 000000000000..fe58a6a32b27
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc64.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc8.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc8.yaml
new file mode 100644
index 000000000000..fe58a6a32b27
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp4_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc128.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc128.yaml
new file mode 100644
index 000000000000..a4a4fe28c7fb
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc128.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1216
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc16.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..397565e15b50
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc256.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc256.yaml
new file mode 100644
index 000000000000..686db04f1fbd
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc256.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1344
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc32.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..397565e15b50
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc4.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..397565e15b50
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc64.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..397565e15b50
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc64.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc8.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..397565e15b50
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/1k1k_tp8_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 1152
+max_seq_len: 2068
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc128.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc128.yaml
new file mode 100644
index 000000000000..ace419c0d81e
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc128.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8384
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc16.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc16.yaml
new file mode 100644
index 000000000000..a0f2de5fec79
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc256.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc256.yaml
new file mode 100644
index 000000000000..3c812ea3e963
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc256.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8512
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc32.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc32.yaml
new file mode 100644
index 000000000000..a0f2de5fec79
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc4.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc4.yaml
new file mode 100644
index 000000000000..a0f2de5fec79
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc64.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc64.yaml
new file mode 100644
index 000000000000..06f600c1cd90
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc64.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc8.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc8.yaml
new file mode 100644
index 000000000000..a0f2de5fec79
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp4_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc128.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc128.yaml
new file mode 100644
index 000000000000..5334ed3cf545
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc128.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8384
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc16.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..382a3c9045da
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc16.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc256.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc256.yaml
new file mode 100644
index 000000000000..639fdde94a18
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc256.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8512
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc32.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..382a3c9045da
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc32.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc4.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..382a3c9045da
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc4.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc64.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..930a6253084f
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc64.yaml
@@ -0,0 +1,22 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: true
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: CUTLASS
+attention_dp_config:
+  batching_wait_iters: 0
+  enable_balance: true
+  timeout_iters: 60
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc8.yaml b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..382a3c9045da
--- /dev/null
+++ b/examples/configs/database/nvidia/DeepSeek-R1-0528-FP4-v2/B200/8k1k_tp8_conc8.yaml
@@ -0,0 +1,18 @@
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 512
+enable_attention_dp: false
+print_iter_log: true
+kv_cache_config:
+  dtype: fp8
+  free_gpu_memory_fraction: 0.8
+  enable_block_reuse: false
+stream_interval: 10
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 8320
+max_seq_len: 9416
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc16.yaml
new file mode 100644
index 000000000000..1d4df970107d
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc32.yaml
new file mode 100644
index 000000000000..7d65f54710bf
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc4.yaml
new file mode 100644
index 000000000000..ca850a775852
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc64.yaml
new file mode 100644
index 000000000000..345b0e501388
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc8.yaml
new file mode 100644
index 000000000000..5fa5e373d2e5
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp1_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc16.yaml
new file mode 100644
index 000000000000..7b392ada8d6b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc32.yaml
new file mode 100644
index 000000000000..e8212dd139ff
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc4.yaml
new file mode 100644
index 000000000000..ab22a7baf682
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc64.yaml
new file mode 100644
index 000000000000..3f8265048033
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc8.yaml
new file mode 100644
index 000000000000..b07960f33ddd
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp2_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc16.yaml
new file mode 100644
index 000000000000..e078ea3d6d32
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc32.yaml
new file mode 100644
index 000000000000..15f5a3ca501d
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc4.yaml
new file mode 100644
index 000000000000..cdbb40a3eb47
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc64.yaml
new file mode 100644
index 000000000000..c5854b6dafb1
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc8.yaml
new file mode 100644
index 000000000000..0ac44311750b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp4_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..a18faa262246
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..4ce42b3ce83f
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..966138c163c4
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..a322f0681d28
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..644d2dabb489
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k1k_tp8_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc16.yaml
new file mode 100644
index 000000000000..31544aa9f4f6
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc32.yaml
new file mode 100644
index 000000000000..ec0ea7b2ba31
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc4.yaml
new file mode 100644
index 000000000000..249b14723ff3
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc64.yaml
new file mode 100644
index 000000000000..21de3414a842
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc8.yaml
new file mode 100644
index 000000000000..315b1add429c
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp1_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc16.yaml
new file mode 100644
index 000000000000..56e1b648bd78
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc32.yaml
new file mode 100644
index 000000000000..4e02fe671b4e
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc4.yaml
new file mode 100644
index 000000000000..4bc360839a97
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc64.yaml
new file mode 100644
index 000000000000..584fb5ae1aee
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc8.yaml
new file mode 100644
index 000000000000..6ab46126d56b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp2_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc16.yaml
new file mode 100644
index 000000000000..ef539d3befa9
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc32.yaml
new file mode 100644
index 000000000000..40dc75208448
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc4.yaml
new file mode 100644
index 000000000000..3e0f48e7e1bf
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc64.yaml
new file mode 100644
index 000000000000..2e3721c712dc
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc8.yaml
new file mode 100644
index 000000000000..098e7ec388cf
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp4_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc16.yaml
new file mode 100644
index 000000000000..45d77f70bd23
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc32.yaml
new file mode 100644
index 000000000000..9436b079590a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc4.yaml
new file mode 100644
index 000000000000..a2917bfd5b19
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc64.yaml
new file mode 100644
index 000000000000..702d3bc00cb1
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc8.yaml
new file mode 100644
index 000000000000..c0b90314c3d0
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/1k8k_tp8_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc16.yaml
new file mode 100644
index 000000000000..31544aa9f4f6
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc32.yaml
new file mode 100644
index 000000000000..ec0ea7b2ba31
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc4.yaml
new file mode 100644
index 000000000000..249b14723ff3
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc64.yaml
new file mode 100644
index 000000000000..21de3414a842
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc8.yaml
new file mode 100644
index 000000000000..315b1add429c
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp1_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc16.yaml
new file mode 100644
index 000000000000..56e1b648bd78
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc32.yaml
new file mode 100644
index 000000000000..4e02fe671b4e
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc4.yaml
new file mode 100644
index 000000000000..4bc360839a97
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc64.yaml
new file mode 100644
index 000000000000..584fb5ae1aee
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc8.yaml
new file mode 100644
index 000000000000..6ab46126d56b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp2_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc16.yaml
new file mode 100644
index 000000000000..ef539d3befa9
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc32.yaml
new file mode 100644
index 000000000000..40dc75208448
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc4.yaml
new file mode 100644
index 000000000000..3e0f48e7e1bf
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc64.yaml
new file mode 100644
index 000000000000..2e3721c712dc
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc8.yaml
new file mode 100644
index 000000000000..098e7ec388cf
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp4_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..45d77f70bd23
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc16.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..9436b079590a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc32.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..a2917bfd5b19
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc4.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..702d3bc00cb1
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc64.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..c0b90314c3d0
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/B200/8k1k_tp8_conc8.yaml
@@ -0,0 +1,22 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+  NCCL_GRAPH_REGISTER: 0
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: fp8
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+print_iter_log: true
+stream_interval: 20
+num_postprocess_workers: 4
+moe_config:
+  backend: TRTLLM
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc16.yaml
new file mode 100644
index 000000000000..2eea897e2f58
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc32.yaml
new file mode 100644
index 000000000000..1a0d44fb2778
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc4.yaml
new file mode 100644
index 000000000000..82662456f032
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc64.yaml
new file mode 100644
index 000000000000..57d8e2ada29b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc8.yaml
new file mode 100644
index 000000000000..87e34788d7c6
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp1_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc16.yaml
new file mode 100644
index 000000000000..57b4b87fc72d
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc32.yaml
new file mode 100644
index 000000000000..0d796e475135
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc4.yaml
new file mode 100644
index 000000000000..f6c41d8bbdf7
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc64.yaml
new file mode 100644
index 000000000000..fdec025db8b9
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc8.yaml
new file mode 100644
index 000000000000..8565e82e3623
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp2_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc16.yaml
new file mode 100644
index 000000000000..4773067517e2
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc32.yaml
new file mode 100644
index 000000000000..5e0d27c5ead3
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc4.yaml
new file mode 100644
index 000000000000..9b135c0a3215
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc64.yaml
new file mode 100644
index 000000000000..6874784b9f29
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc8.yaml
new file mode 100644
index 000000000000..cc1d2d8ac986
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp4_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..f7e46b17a369
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..1b1b874c3e1b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..28a7f3d17c69
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..8036e7439962
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..12289904eda8
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k1k_tp8_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 2068
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc16.yaml
new file mode 100644
index 000000000000..7ccdc4ae1146
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc32.yaml
new file mode 100644
index 000000000000..ea6a93ba64bb
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc4.yaml
new file mode 100644
index 000000000000..a0149f2ab5c5
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc64.yaml
new file mode 100644
index 000000000000..3ae56a300ad6
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc8.yaml
new file mode 100644
index 000000000000..c18bc3c7581c
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp1_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc16.yaml
new file mode 100644
index 000000000000..e88b4e05fe8a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc32.yaml
new file mode 100644
index 000000000000..95b8e207330d
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc4.yaml
new file mode 100644
index 000000000000..c35b691a81c5
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc64.yaml
new file mode 100644
index 000000000000..ce0f7c27577c
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc8.yaml
new file mode 100644
index 000000000000..344166bc325b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp2_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc16.yaml
new file mode 100644
index 000000000000..4f895199b1f2
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc32.yaml
new file mode 100644
index 000000000000..ca549de3d28a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc4.yaml
new file mode 100644
index 000000000000..b87044bbc0ea
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc64.yaml
new file mode 100644
index 000000000000..9af104970e78
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc8.yaml
new file mode 100644
index 000000000000..7440c3fcb7e7
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp4_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc16.yaml
new file mode 100644
index 000000000000..b1d8a6eead07
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc32.yaml
new file mode 100644
index 000000000000..f8c7fec13a7a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc4.yaml
new file mode 100644
index 000000000000..f9cb8feb6956
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc64.yaml
new file mode 100644
index 000000000000..a9124d700757
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc8.yaml
new file mode 100644
index 000000000000..7c2507ace7aa
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/1k8k_tp8_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc16.yaml
new file mode 100644
index 000000000000..7ccdc4ae1146
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc32.yaml
new file mode 100644
index 000000000000..ea6a93ba64bb
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc4.yaml
new file mode 100644
index 000000000000..a0149f2ab5c5
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc64.yaml
new file mode 100644
index 000000000000..3ae56a300ad6
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc8.yaml
new file mode 100644
index 000000000000..c18bc3c7581c
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp1_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc16.yaml
new file mode 100644
index 000000000000..e88b4e05fe8a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc32.yaml
new file mode 100644
index 000000000000..95b8e207330d
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc4.yaml
new file mode 100644
index 000000000000..c35b691a81c5
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc64.yaml
new file mode 100644
index 000000000000..ce0f7c27577c
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc8.yaml
new file mode 100644
index 000000000000..344166bc325b
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp2_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 2
+moe_expert_parallel_size: 2
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc16.yaml
new file mode 100644
index 000000000000..4f895199b1f2
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc32.yaml
new file mode 100644
index 000000000000..ca549de3d28a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc4.yaml
new file mode 100644
index 000000000000..b87044bbc0ea
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc64.yaml
new file mode 100644
index 000000000000..9af104970e78
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc8.yaml
new file mode 100644
index 000000000000..7440c3fcb7e7
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp4_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 4
+moe_expert_parallel_size: 4
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc16.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc16.yaml
new file mode 100644
index 000000000000..b1d8a6eead07
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc16.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 16
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc32.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc32.yaml
new file mode 100644
index 000000000000..f8c7fec13a7a
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc32.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 32
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc4.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc4.yaml
new file mode 100644
index 000000000000..f9cb8feb6956
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc4.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 4
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc64.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc64.yaml
new file mode 100644
index 000000000000..a9124d700757
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc64.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 64
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc8.yaml b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc8.yaml
new file mode 100644
index 000000000000..7c2507ace7aa
--- /dev/null
+++ b/examples/configs/database/openai/gpt-oss-120b/H200/8k1k_tp8_conc8.yaml
@@ -0,0 +1,21 @@
+env_overrides:
+  TRTLLM_ENABLE_PDL: 1
+cuda_graph_config:
+  enable_padding: true
+  max_batch_size: 8
+enable_attention_dp: false
+kv_cache_config:
+  dtype: auto
+  enable_block_reuse: false
+  free_gpu_memory_fraction: 0.85
+moe_config:
+  backend: TRITON
+num_postprocess_workers: 4
+print_iter_log: true
+stream_interval: 20
+tensor_parallel_size: 8
+moe_expert_parallel_size: 8
+trust_remote_code: true
+backend: pytorch
+max_num_tokens: 20000
+max_seq_len: 9236
diff --git a/examples/models/core/qwen/README.md b/examples/models/core/qwen/README.md
index 1d3d97b26796..8bdd9e4bf5d5 100644
--- a/examples/models/core/qwen/README.md
+++ b/examples/models/core/qwen/README.md
@@ -747,7 +747,7 @@ We maintain YAML configuration files with recommended performance settings in th
 
 ```shell
 TRTLLM_DIR=/app/tensorrt_llm # change as needed to match your environment
-EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/qwen3.yaml
+EXTRA_LLM_API_FILE=${TRTLLM_DIR}/examples/configs/curated/qwen3.yaml
 ```
 
 #### trtllm-serve
@@ -778,7 +778,7 @@ For example, you can launch a single context server on port 8001 with:
 ```bash
 export TRTLLM_USE_UCX_KVCACHE=1
 export TRTLLM_DIR=/app/tensorrt_llm
-export EXTRA_LLM_API_FILE="${TRTLLM_DIR}/examples/configs/qwen3-disagg-prefill.yaml"
+export EXTRA_LLM_API_FILE="${TRTLLM_DIR}/examples/configs/curated/qwen3-disagg-prefill.yaml"
 
 trtllm-serve Qwen3-30B-A3B/ --port 8001 --extra_llm_api_options ${EXTRA_LLM_API_FILE} &> output_ctx &
 ```
@@ -788,7 +788,7 @@ And you can launch two generation servers on port 8002 and 8003 with:
 ```bash
 export TRTLLM_USE_UCX_KVCACHE=1
 export TRTLLM_DIR=/app/tensorrt_llm
-export EXTRA_LLM_API_FILE="${TRTLLM_DIR}/examples/configs/qwen3.yaml"
+export EXTRA_LLM_API_FILE="${TRTLLM_DIR}/examples/configs/curated/qwen3.yaml"
 
 for port in {8002..8003}; do \
 trtllm-serve Qwen3-30B-A3B/ --port ${port} --extra_llm_api_options ${EXTRA_LLM_API_FILE} &> output_gen_${port} & \
diff --git a/scripts/generate_config_table.py b/scripts/generate_config_table.py
new file mode 100644
index 000000000000..2d423c0811fc
--- /dev/null
+++ b/scripts/generate_config_table.py
@@ -0,0 +1,169 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+import os
+import sys
+from collections import defaultdict
+from pathlib import Path
+
+from examples.configs.database.database import DATABASE_LIST_PATH, RecipeList
+
+SCRIPT_DIR = Path(__file__).parent.resolve()
+REPO_ROOT = SCRIPT_DIR.parent
+MODEL_INFO = {
+    "deepseek-ai/DeepSeek-R1-0528": {
+        "display_name": "DeepSeek-R1",
+        "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528",
+    },
+    "nvidia/DeepSeek-R1-0528-FP4-v2": {
+        "display_name": "DeepSeek-R1 (NVFP4)",
+        "url": "https://huggingface.co/nvidia/DeepSeek-R1-0528-FP4-v2",
+    },
+    "openai/gpt-oss-120b": {
+        "display_name": "gpt-oss-120b",
+        "url": "https://huggingface.co/openai/gpt-oss-120b",
+    },
+}
+
+LOW_LATENCY_CONCURRENCY_THRESHOLD = 8
+HIGH_THROUGHPUT_CONCURRENCY_THRESHOLD = 32
+
+
+def generate_rst(yaml_path, output_file=None):
+    """Generate RST table from YAML config database.
+
+    Args:
+        yaml_path: Path to lookup.yaml (str or Path)
+        output_file: Optional output file path. If None, prints to stdout.
+    """
+    recipe_list = RecipeList.from_yaml(Path(yaml_path))
+
+    # Group by model -> (gpu, isl, osl) -> list of recipes
+    model_groups = defaultdict(lambda: defaultdict(list))
+    for recipe in recipe_list:
+        key = (recipe.gpu, recipe.isl, recipe.osl)
+        model_groups[recipe.model][key].append(recipe)
+
+    lines = []
+
+    # Include note_sections.rst at the top (relative include for Sphinx)
+    lines.append(".. include:: note_sections.rst")
+    lines.append("   :start-after: .. start-note-traffic-patterns")
+    lines.append("   :end-before: .. end-note-traffic-patterns")
+    lines.append("")
+
+    sorted_models = sorted(model_groups.keys())
+
+    for model in sorted_models:
+        lines.append(f".. start-{model}")
+        lines.append("")
+
+        if model in MODEL_INFO:
+            info = MODEL_INFO[model]
+            title_text = f"`{info['display_name']} <{info['url']}>`_"
+        else:
+            title_text = model
+
+        lines.append(f".. _{model}:")
+        lines.append("")
+        lines.append(title_text)
+        lines.append("^" * len(title_text))
+        lines.append("")
+
+        lines.append(".. list-table::")
+        lines.append("   :width: 100%")
+        lines.append("   :header-rows: 1")
+        lines.append("   :widths: 12 15 15 13 20 25")
+        lines.append("")
+        lines.append("   * - GPU")
+        lines.append("     - Performance Profile")
+        lines.append("     - ISL / OSL")
+        lines.append("     - Concurrency")
+        lines.append("     - Config")
+        lines.append("     - Command")
+
+        subgroups = model_groups[model]
+        sorted_keys = sorted(
+            subgroups.keys(), key=lambda k: (str(k[0]), int(k[1] or 0), int(k[2] or 0))
+        )
+
+        for key in sorted_keys:
+            entries = subgroups[key]
+            entries.sort(key=lambda x: x.concurrency)
+            n = len(entries)
+
+            for idx, entry in enumerate(entries):
+                gpu = entry.gpu
+                num_gpus = entry.num_gpus
+                gpu_display = f"{num_gpus}x{gpu}" if num_gpus and num_gpus > 1 else gpu
+                isl = entry.isl
+                osl = entry.osl
+                conc = entry.concurrency
+                config_path = entry.config_path
+
+                if n == 1:
+                    if conc <= LOW_LATENCY_CONCURRENCY_THRESHOLD:
+                        profile = "Low Latency"
+                    elif conc >= HIGH_THROUGHPUT_CONCURRENCY_THRESHOLD:
+                        profile = "High Throughput"
+                    else:
+                        profile = "Balanced"
+                elif idx == 0:
+                    profile = "Min Latency"
+                elif idx == n - 1:
+                    profile = "Max Throughput"
+                elif idx in ((n - 1) // 2, n // 2):
+                    profile = "Balanced"
+                elif idx < n // 2:
+                    profile = "Low Latency"
+                else:
+                    profile = "High Throughput"
+
+                full_config_path = config_path
+                command = f"trtllm-serve {model} --extra_llm_api_options ${{TRTLLM_DIR}}/{full_config_path}"
+
+                config_filename = os.path.basename(full_config_path)
+
+                github_url = f"https://github.com/NVIDIA/TensorRT-LLM/blob/main/{full_config_path}"
+                config_link = f"`{config_filename} <{github_url}>`_"
+
+                lines.append(f"   * - {gpu_display}")
+                lines.append(f"     - {profile}")
+                lines.append(f"     - {isl} / {osl}")
+                lines.append(f"     - {conc}")
+                lines.append(f"     - {config_link}")
+                lines.append(f"     - ``{command}``")
+
+        lines.append("")
+        lines.append(f".. end-{model}")
+        lines.append("")
+
+    output_text = "\n".join(lines)
+    if output_file:
+        with open(output_file, "w") as f:
+            f.write(output_text)
+        print(f"Generated table written to: {output_file}", file=sys.stderr)
+    else:
+        print(output_text)
+
+
+if __name__ == "__main__":
+    yaml_path = DATABASE_LIST_PATH
+    if not yaml_path.exists():
+        print(f"Error: YAML file not found at {yaml_path}", file=sys.stderr)
+        sys.exit(1)
+    output_path = REPO_ROOT / "docs/source/deployment-guide/config_table.rst"
+    generate_rst(yaml_path, output_file=output_path)
diff --git a/tensorrt_llm/llmapi/llm_args.py b/tensorrt_llm/llmapi/llm_args.py
index 9f154c53f63e..e62d7f7d0399 100644
--- a/tensorrt_llm/llmapi/llm_args.py
+++ b/tensorrt_llm/llmapi/llm_args.py
@@ -1962,9 +1962,17 @@ class BaseLlmArgs(StrictBaseModel):
     env_overrides: Optional[Dict[str, str]] = Field(
         default=None,
         description=
-        "[EXPERIMENTAL] Environment variable overrides. NOTE: import-time-cached env vars in the code won’t update unless the code fetches them from os.environ on demand.",
+        "[EXPERIMENTAL] Environment variable overrides. NOTE: import-time-cached env vars in the code won't update unless the code fetches them from os.environ on demand.",
         status="prototype")
 
+    @field_validator('env_overrides', mode='before')
+    @classmethod
+    def coerce_env_overrides_to_str(cls, v):
+        """Coerce env_overrides values to strings for os.environ compatibility."""
+        if v is None:
+            return v
+        return {str(k): str(val) for k, val in v.items()}
+
     _parallel_config: Optional[_ParallelConfig] = PrivateAttr(default=None)
     _model_format: Optional[_ModelFormatKind] = PrivateAttr(default=None)
     _speculative_model: Optional[str] = PrivateAttr(default=None)
diff --git a/tests/unittest/llmapi/test_config_database.py b/tests/unittest/llmapi/test_config_database.py
new file mode 100644
index 000000000000..72dfce770f56
--- /dev/null
+++ b/tests/unittest/llmapi/test_config_database.py
@@ -0,0 +1,64 @@
+# SPDX-FileCopyrightText: Copyright (c) 2024-2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""L0 tests for validating config database YAML files against TorchLlmArgs."""
+
+from pathlib import Path
+from unittest.mock import Mock, patch
+
+import pytest
+import yaml
+
+from tensorrt_llm.llmapi.llm_args import TorchLlmArgs, update_llm_args_with_extra_dict
+
+CONFIG_ROOT = Path(__file__).parents[3] / "examples" / "configs"
+DATABASE_DIR = CONFIG_ROOT / "database"
+
+DATABASE_CONFIGS = (
+    [c for c in DATABASE_DIR.rglob("*.yaml") if c.name != "lookup.yaml"]
+    if DATABASE_DIR.exists()
+    else []
+)
+
+
+@pytest.fixture(autouse=True)
+def mock_gpu_environment():
+    """Mock GPU functions for CPU-only test execution."""
+    mock_props = Mock()
+    mock_props.major = 8
+
+    with patch("torch.cuda.device_count", return_value=8):
+        with patch("torch.cuda.get_device_properties", return_value=mock_props):
+            with patch("torch.cuda.is_available", return_value=True):
+                yield
+
+
+def get_config_id(config_path: Path) -> str:
+    return str(config_path.relative_to(DATABASE_DIR))
+
+
+@pytest.mark.part0
+@pytest.mark.parametrize("config_path", DATABASE_CONFIGS, ids=get_config_id)
+def test_config_validates_against_llm_args(config_path: Path):
+    with open(config_path) as f:
+        config_dict = yaml.safe_load(f) or {}
+
+    base_args = TorchLlmArgs(model="dummy/model", skip_tokenizer_init=True)
+    merged = update_llm_args_with_extra_dict(base_args.model_dump(), config_dict)
+    TorchLlmArgs(**merged)
+
+
+@pytest.mark.part0
+def test_database_config_count():
+    assert len(DATABASE_CONFIGS) > 0, "No database config files found"
diff --git a/tests/unittest/tools/test_generate_config_table.py b/tests/unittest/tools/test_generate_config_table.py
new file mode 100644
index 000000000000..a2dcf66783fa
--- /dev/null
+++ b/tests/unittest/tools/test_generate_config_table.py
@@ -0,0 +1,66 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import sys
+import tempfile
+import unittest
+
+# Add scripts directory to path
+REPO_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), "../../.."))
+SCRIPTS_DIR = os.path.join(REPO_ROOT, "scripts")
+sys.path.insert(0, SCRIPTS_DIR)
+
+from generate_config_table import generate_rst  # noqa: E402
+
+
+class TestConfigTableSync(unittest.TestCase):
+    def test_config_table_sync(self):
+        """Test that the config_table.rst file is synchronized with the lookup.yaml database.
+
+        Ensures that the RST file is up-to-date with the YAML database.
+        """
+        if generate_rst is None:
+            self.skipTest("generate_config_table not available")
+
+        # Define paths
+        yaml_path = os.path.join(REPO_ROOT, "examples/configs/database/lookup.yaml")
+        rst_path = os.path.join(REPO_ROOT, "docs/source/deployment-guide/config_table.rst")
+
+        # Ensure files exist
+        self.assertTrue(os.path.exists(yaml_path), f"YAML file not found: {yaml_path}")
+        self.assertTrue(os.path.exists(rst_path), f"RST file not found: {rst_path}")
+
+        # Read existing RST content
+        with open(rst_path, "r") as f:
+            existing_content = f.read()
+
+        # Generate new RST content
+        with tempfile.NamedTemporaryFile(mode="w+", delete=True) as tmp:
+            generate_rst(yaml_path, output_file=tmp.name)
+            tmp.seek(0)
+            generated_content = tmp.read()
+
+        # Compare content
+        self.assertEqual(
+            existing_content.strip(),
+            generated_content.strip(),
+            "config_table.rst is not synchronized with lookup.yaml. "
+            "Please run 'python3 scripts/generate_config_table.py' from the repo root to update it.",
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()