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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [CI][ROCm] Add AMD HC kernel tests for Qwen4 Exp and mirror the B200 HC step
- [CI][ROCm] Mirror BF16x3 Router GEMM Kernels on MI355
- [ROCm][CI] Run XPress on MI355 with a GEMM latent MLP and mirror its nightly
- [Feature]: EAGLE-3 / DSpark support (aux hidden states) for Lfm3ForCausalLM
- [ROCm][Model] Run Qwen3.5 on the eager breakable-graph model on MI355X
- [ROCm][CI] Support prefill context parallel for GLM-5.2 sparse MLA and mirror LM Eval PCP
- [Model] Add EAGLE-3 / DSpark support (aux hidden states) for Lfm3ForCausalLM
- [ROCm] Use 128x128x128 MXFP8 tiles for large gfx950 prefill GEMMs
- [Bugfix][Scheduler] Bound cache-aware admission probe by request reuse limit
- [Bug]: /v1/audio/transcriptions and /translations reject stream_include_usage=false (form strings checked before bool coercion)
- Docs
- Python not yet supported