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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: `--shutdown-timeout` never drains in-flight requests when `--api-server-count > 1`: the launcher stops the API servers before the engines, so every in-flight request fails with EngineDeadError and the engines then drain for nobody
- [ROCm][Kimi-K3] Mono MoE for decode steps the specialized decode does not take
- [Feature][Observability] Add tool-calling metrics
- [RustFrontend] Add shm mm cache support for rust frontend
- [Model Loader] Skip unused tensors with InstantTensor
- [Core][KV Cache] Move KV cache allocation policy out of KVCacheSpecs (RFC #42449 part 2)
- [Model] Enable LoRA support for tower and connector in MiDashengLM
- [Bug]: HiSparse with --enable-dbo crashes in dbo_register_recv_hook; construction-time workspace reservations only size ubatch slot 0
- [Simulator] Timing, recording, disaggregated prefill and configuration search for simulated execution
- [Bug]: [KV Offload][P2P] Consumer frees load slots on abort-ack timeout while the producer's WRITE may still land
- Docs
- Python not yet supported