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
- test(sampling_params): add unit tests for from_optional and logit_bias validation
- [Usage]: Does num_speculative_tokens>1 reuse only layer0 for multi-layer MTP (qwen3_5_mtp)?
- [RFC]: support rl feature : weight checker
- [Bugfix][V1] Gate async-scheduling output-token-id repair on a declared logitsproc capability
- [Bugfix][Optimization] Zero packed KV blocks once from the backing sorage base
- [Bug]: GPTQ MoE with `group_size: -1` crashes in `MoeWNA16Method.create_weights` (reached via the ROCm Marlin fallback)
- [Performance][MRV2] Use uint16 for penalty output counts
- [Platform][MRV2] Make MRV2 Components Pluggable
- feat(rust): decode routed expert EngineCore output
- feat(rust): return routed experts through gRPC
- Docs
- Python not yet supported