vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported13 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Use a different prompt for benchmark_serving.py's test prompt
- [Bug]: vLLM doesn't release memory on deletion of the LLM runner with VLLM_ENABLE_V1_MULTIPROCESSING=0
- Add tuned fused_moe configs for Qwen3-30B-A3B
- Remove `build_for_cudagraph_capture` method and use `build` directly in attention metadata builders
- [Quantization] Use CompressedTensors utils matching utils, support module ignores
- [ROCm][FEAT] Integrate AITER tgemm.
- Use Numpy array for sampled_token_ids
- [Bug]: Running on AMD Epyc 9654 (CPU Only) always tries to use intel_extension_for_pytorch and crashes.
- [Bug]: vLLM Runtime Fails to Honor Context Cancellation During Streaming
- draft cp&sp
- Docs
- Python not yet supported