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
- [Hybrid] Round the attention block size up to prefix_match_unit
- [Feature]: Support LiquidAI LFM3 models
- [Benchmark] Add small-M NVFP4 linear benchmark across vLLM kernels
- [RFC]: Direct layerwise and chunk-wise KV WRITE with Mooncake
- [Bugfix][Scheduler] Cap the async-load spec margin at max_model_len
- [Bug]: EAGLE3 draft acceptance collapses after the first position with --block-size > 128 (MiniMax-M3, GB200)
- [Bugfix][Rust Frontend] Serve SageMaker /ping and mark /invocations unsupported
- [Bugfix][NIXL] Preserve packed-row aliases during region dedup
- [ROCm][Perf] Add MI325X FP8 fused MoE configs for Qwen3.8-Flash-Next
- [Frontend] Add --enable-log-request-ids-only
- Docs
- Python not yet supported