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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Feature] Batch-invariant support for speculative decoding
- [Bug]: Marlin MoE output depends on equivalent within-expert token ordering
- [Metrics] Report shared-prefix tokens lost to a missing sparse-retention checkpoint
- [Bug]: Kimi-K3 CUDA graph capture silently corrupts output at batch=1; three distinct failure modes across cudagraph modes.
- Streaming input: per-chunk arrival cost is O(cumulative prompt), making long multimodal sessions quadratic
- [Bugfix][Core] Count SWA in-flight KV once per pool, not per request (DeepSeek-V4-Flash-0731 #51041)
- [Bugfix][Hardware][NVIDIA] Fix DSV4 sparse MLA spec-decode shapes on SM120 FlashInfer path
- [Bugfix][Frontend] Keep vLLM MCP x-session-id over client headers
- [Bug]: Under async scheduling, stateful logits processors that read output_token_ids values see -1 placeholders — logitsprocs_need_output_token_ids only counts CLI custom processors
- [Bugfix] Fix use_existing_torch.py stripping unrelated torch-* packages
- Docs
- Python not yet supported