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 supported13 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Fix] Add SIGINT handler for openai
- [Bug]: Qwen3-VL-8B-Instruct has accuracy issue - Multi modal accuracy issue
- [Performance]: If the next request is sent immediately after the previous one finishes, its TTFT will be relatively small; if the next request is sent 10 seconds after the previous one ends, its TTFT will be relatively large.
- [perf]skip embedding check if have computed
- [Bug]: After upgrade to 0.11.2, vllm crashs with Qwen3.
- [Installation]: cuda error when install from source
- Add `/pause/step` fast pause + barrier for online weight update
- [perf][MLA] Fuse RoPE/FP8 quantization/Q write using mla_rope_quantize_fp8
- [perf] Fused operator SplitMrope used in the Qwen2.5-Omni-7B model
- [Model] Enable LoRA support for LLaVA family
- Docs
- Python not yet supported