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 supported36 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Perf] Reduce Triton recompiles for multimodal attention ranges
- [BugFix] Sample uniform routing-sim experts without replacement
- [RFC]: Return extracted hidden states in the generation response
- [Bug]: Gemma-4-26B-A4B-it MoE fails to deploy on XPU backend (vLLM v0.21.0)
- [Bug]: Spurious EngineDeadError traceback logged during graceful shutdown (AsyncLLM.shutdown cancels output_handler after engine teardown)
- [Bugfix][Model] Align Ministral3 scaling config with Transformers
- [XPU] Fix Dockerfile.xpu: add /opt/venv/lib to LD_LIBRARY_PATH
- [Bugfix] Emit DeepSeek V3/V3.1 tool calls that arrive in a single streaming delta
- [Bugfix][LoRA] Avoid block_n=128 for lora_expand on Hopper
- [Bug]: Runtime-varying tl.constexpr params force Triton kernel recompiles on the serving path (unified attention MAX_MM_RANGES et al.)
- Docs
- Python not yet supported