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 supported60 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [KV cache] Add an mxfp4 kv_cache_dtype
- [Perf][DSv4.1] Improve the TP4 concurrency-1 specialized decode
- [Bugfix][torch.compile] Read traced source files as UTF-8 when hashing
- [KV Connector][NIXL] Add connector checkpoint hooks and restore NIXL P/D workers independently
- [RFC]: Bounded, retryable NCCL weight updates
- [Perf][Core] Admit async KV loads in linear time per scheduling step
- [Core] Packed KV planner: pick the block width that reserves the fewest bytes
- [ROCm][Bugfix] Route padded MoE rows to expert 0 with zero weight on the AITER a4w4 path
- [Bug]: TimeoutError: RPC call to sample_tokens timed out.
- [Bugfix][KV Offload] SimpleCPUOffload: remove the disk tier's file on graceful shutdown
- Docs
- Python not yet supported