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 supported26 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: DeepGemmExperts FP8 MoE drops swiglu_limit (DeepSeek-V4-Flash-Base produces glitch tokens)
- [Feature]: Opt-in: stop caching KV blocks after thinking start tokens
- [Bugfix] Add Pixtral processor image token properties
- [Bug]: v1 Scheduler: preempted requests lose re-admission priority in PriorityRequestQueue, causing redundant recompute
- [Bug]: Mixed INT4/INT8 GPTQ MoE models crash on initialization (AssertionError in fused_marlin_moe)
- [Bug]: Gemma4 + MTP speculative decoding drops first tool-call arguments in streaming multi-tool auto-tool-choice
- [Bug]: AsyncLLM silent-hangs on multimodal pooling requests when default max_num_batched_tokens too small (L4); sync LLM works on same hardware
- Stale padded request metadata can misclassify Mamba CUDA graph rows
- [CI Failure]: mi300_4: RayExecutorV2 (4 GPUs)
- [CI Failure]: mi300_1: Multi-Modal Models (Extended Generation 3)
- Docs
- Python not yet supported