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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: GLM-5.3-Flash with --mm-encoder-tp-mode data gives wrong image and video answers (images look mixed up), TP=2 on B200, v0.31.0
- [ROCm][Perf][DSV4] Keep one decoder-replay row on unfinished prefills on gfx942
- [Feature][Kernel] Triton autotune block fp8
- [ROCm][Perf] Request fp32 output from the AITER router GEMM (otype)
- [Feature][Kernel] Triton autotune fused MoE
- [Fix] Stop MRV2 prompt logprobs at the end of the prompt
- [ROCm][Perf][Kimi-K3] Specialized decode: verify steps of 17-32 rows
- [Bug]: v0.25.0 picks FlashInfer mnnvl allreduce when multicast bind fails, then crashes with illegal memory access (2x H100)
- [Perf][Multimodal] Use FA4 for long-sequence ViT attention on Hopper
- [Bugfix] Re-probe the tier for new requests instead of reusing a cached miss in tiering AsyncLookupManager
- Docs
- Python not yet supported