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 supported31 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Rust Frontend] Avoid redundant decode per token in incremental detokenizer
- [Attention] Occupancy-gated 3D segmented decode for multi-query (diffusion-LM canvas) over long KV
- [Bugfix] Avoid MLA decode shared memory OOR
- [Build] Update vllm to point to vllm-project/flash-attention commit that builds FA3 with torch stable API.
- [ROCm][CI] Relax fused layernorm quant test tolerances for one-ULP outliers
- [Bugfix][Frontend] Fix reasoning-end detection to check prompt tail only
- [Bug]: GLM5.2 RuntimeError: The expanded size of the tensor (10210) must match the existing size (16384) at non-singleton dimension 1. Target sizes: [12, 10210]. Tensor sizes: [12, 16384]
- [WIP] add helion nvfp4 backend for batch_size=1(gemv)
- docs: document NIXL KV connector transfer metrics aggregation logic
- [Bugfix] Skip PP sampled-token broadcast on KV producer
- Docs
- Python not yet supported