[https://nvbugs/6018046][fix] Drop max_batch_size 32→8 for throughput_pp4_mtp (matching throughput_bs8_mtp), l#14201

tensorrt-cicd wants to merge 1 commit into

NVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

tensorrt-cicd:repair-bot-bug6018046tensorrt-cicd/TensorRT-LLM:repair-bot-bug6018046Copy head branch name to clipboard