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
- [Perf][HYV4] Fuse iHC boundaries across decoder layers
- [Attention] Enable FA4 CUDA graphs for adaptive verification
- [KV Offload] Add compact filesystem files
- [Bug]: `VLLM_DEBUG_MFU_METRICS=1` shuts the server down on the first request (msgspec `ValidationError` on `context_breakdown`)
- [Bugfix][CPU] Default to V1 model runner on CPU backend (#58565)
- [Bugfix][Metrics] Keep per-request records out of the debug wire payload
- Docs/nixl kv metrics aggregation semantics
- [Bugfix][KV Offload] Enable partial-tail offload for hybrid models under DCP
- [CI] Raise H200 MIG CUDAGraph timeout 30m -> 40m
- [Bugfix][Frontend] Release admission slot on cancelled submission
- Docs
- Python not yet supported