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]: V2 sampler warmup misses explicit-seed native path when FlashInfer is enabled
- [RFC] Qwen3.8-Flash-Next: fp8_e4m3 KV cache on the QSA path — working patch, one machine, looking for corroboration
- [Rust Frontend][Perf] Fast-path single-token tiktoken decoding
- [Spec Decode][Perf] Support vocab-parallel DSpark top-k for DeepSeek V4
- [Bug] hy_v4 (Hy4-preview): degenerate output with FLASHINFER_MLA_SPARSE_SM120 — the only available sparse backend on SM120
- [Bug]: structured output can sample unconstrained tokens when the draft hand-off goes stale (spec decode + async scheduling + PP)
- [RFC]: Data integrity and I/O liveness for the filesystem KV offload tier
- [Core] Add waiting-queue-informed KV cache eviction
- [Bugfix][Frontend] Normalize null assistant content with tool_calls
- [Distributed] Implement stubbed collective operations for Gloo backend in `DeviceCommunicatorBase`
- Docs
- Python not yet supported