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 supported26 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: Performance regression from v0.16.0 to v0.17.0+ on openai/gpt-oss-120b
- [Spec Decode] Inherit online quantization for draft models
- [Bug]: DeepSeek-V4-Pro H200 DP+EP router dtype mismatch in topk_hash_softplus_sqrt (Long/Int inconsistency)
- [Bug]: QWEN3.5-27B fails to mount LoRA
- [Bug]: ngram speculative decoding default prompt_lookup_min=2 causes tool-call output corruption on Qwen3-class models with structured output (config-only fix: prompt_lookup_min=8)
- [Attention] Move FA3→FA4 upgrade into get_flash_attn_version()
- [Feature] Triton INT4 / INT2 per-token-head KV cache quantization + Prefill + Decode Kernels Per Token Head
- [Bugfix] Fix `NameError` on `AsyncEngineArgs.quantization_config` forward reference
- [vLLM IR] Propagate lowering annotations through source_fn_stack
- [V1][Spec Decode] Avoid O(N*K) membership scan in NgramProposer.batch_propose
- Docs
- Python not yet supported