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
- [Bug]: `ngram_gpu` speculative decoding with xgrammar returns HTTP 500 under concurrent structured-output requests
- [Docs] Document NixlPushConnector for same-node / non-RDMA KV transfer
- [Documentation]: Clarify FileStore vs TCPStore security behavior by executor and version
- [Bug]: OffloadingConnector silently crashes at >64GB CPU RAM with GLM 5.2
- [Perf] Avoid FlashInfer MLA prefill GPU sync
- [Bug]: FP8 on RDNA3/gfx1100 exceeds the 600 s engine-ready timeout on first start — no AMD_Radeon_Graphics tuned configs shipped
- feat(entrypoints): add Gate/Prove ActionBoundary guardrail and Action Ledger exporter for OpenAI server
- [Bug] Streaming input: zero-token continuation crashes EngineCore (Invalid request status: RUNNING) — client-triggerable via temperature>0 chat sessions
- [Doc] Document cache-usage reporting flag and prefix_cache_retention_interval
- feat(core): add ProductionDebtPagedAttentionGate and TechnicalDueDiligenceLedger
- Docs
- Python not yet supported