[LLM] Support for automatic deployment of services, modification of environment variable names #9966
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #9966 +/- ##
============================================
+ Coverage 17.38% 50.56% +33.17%
============================================
Files 752 752
Lines 120314 121062 +748
============================================
+ Hits 20919 61213 +40294
+ Misses 99395 59849 -39546 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Jiang-Jia-Jun
suggested changes
Mar 4, 2025
| export BLOCK_BS=${BLOCK_BS:-"4"} | ||
| export BLOCK_SIZE=${BLOCK_SIZE:-"64"} | ||
| export DTYPE=${DTYPE:-"bfloat16"} | ||
| export USE_CACHE_KV_INT8=${USE_CACHE_KV_INT8:-"0"} # c8 model requires configuration 1 |
| export METRICS_PORT=${METRICS_PORT:-"8722"} | ||
| export INFER_QUEUE_PORT=${INFER_QUEUE_PORT:-"8813"} | ||
| export PUSH_MODE_HTTP_PORT=${PUSH_MODE_HTTP_PORT:-"9965"} | ||
| export HEALTH_HTTP_PORT=${HTTP_PORT:-${HEALTH_HTTP_PORT:-"8110"}} |
llm/server/server/server/utils.py
Outdated
| # Cleanup on failure | ||
| if os.path.exists(temp_tar): | ||
| os.remove(temp_tar) | ||
| raise Exception(f"Failed to get model from {url}, please recheck the model name from https://github.com/PaddlePaddle/PaddleNLP/blob/develop/llm/docs/predict/inference.md") |
There was a problem hiding this comment.
这里的逻辑是，不管用户是什么字符串，都先去尝试？ 然后下载不到抛这个错误吗
| # Check if model_name matches any supported pattern | ||
| if not any(re.match(pattern, model_name) for pattern in supported_patterns): | ||
| raise ValueError( | ||
| f"{model_name} is not in the supported list. Currently supported models: Qwen, Llama, Mixtral, DeepSeek." |
There was a problem hiding this comment.
改成 当前Serving 仅支持 Qwen, LlaMA, Mixtral, DeepSeekv3系列模型，具体Model name参数xx文档
Jiang-Jia-Jun
approved these changes
Mar 6, 2025
ZHUI
reviewed
Mar 6, 2025
| @@ -0,0 +1 @@ | |||
| from server.triton_server import TritonPythonModel No newline at end of file | |||
ZHUI
approved these changes
Mar 6, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Before submitting
testsfolder. If there are codecov issues, please add tests cases first.PR types
PR changes
environment variable modification
add the feature to download models.
Description
In this update, we have adjusted the naming of environment variables to enhance the clarity and consistency of configurations. Below are the specific changes:
GRPC_PORThas been changed toSERVICE_GRPC_PORTHTTP_PORThas been changed toHEALTH_HTTP_PORTMETRICS_PORThas been changed toMETRICS_HTTP_PORTINTER_QUEUE_PORThas been changed toINTER_PROC_PORTPUSH_MODE_HTTP_PORThas been changed toSERVICE_HTTP_PORTThe original environment variables remain compatible, but it is recommended to use the latest ones.
add
download_modelfunction to download the model from the repository.