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 supported37 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: Default eplb num_redundant_experts to minimum valid value (#30075)
- [Perf][Kernel] Opt-in fused QK norm+RoPE, SiLU+FP8, and ShortConv decode kernels for LFM2.5
- fix(uva): fix silent input corruption and remote DoS under GPU CC
- [ROCm][AMD] Kimi-K3 Gap and Roadmap Tracking
- [Bugfix] Fix DeepSeek-V4 reasoning_effort "high" tier and message-level tools preservation
- fix(deepseek_v4): merge consecutive assistant messages in prompt enco…
- [Docs] Fix dead class references in MoE design docs
- [Bug] gdn_attn: cudagraph padding branch guards on num_decodes but slices by batch_size
- [Bugfix] Stream Kimi K3 tool-call arguments incrementally
- [Bug]: runai_streamer multi-node: aux (non-tensor) files are not pulled on peer nodes
- Docs
- Python not yet supported