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 supported52 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Kernel]: _flash_attn_fwd() got an unexpected keyword argument 'dynamic_causal'
- fix(flashinfer): guard trtllm MoE behind x86_64 check
- [Metrics] Skip unknown metric types in `get_metrics_snapshot()`
- [Bugfix] Package example Jinja chat templates in wheels
- [Core] Reserve FlashInfer's persistent workspace before KV cache sizing
- [BugFix] Report correct cached_tokens for disaggregated prefill
- fix: [Bug]: ModelOpt Llama-4 Checkpoints Take 5+ minutes to load
- [Bug]: DeepSeek-V4-Flash fails to start on B300 (SM100) — DeepGEMM sm100_tf32_hc_prenorm_gemm kernel launch fails (invalid argument on CUDA 13, invalid configuration argument on CUDA 12.9)
- MTP + MLA: slot_mapping assert fails with list[dict] from ubatch slicing
- [ROCM][Bugfix] Deferring to TRITON_ATTN backend for Prefix lm models instead of ROCM_ATTN and fixed minor bug in Gemma3 Multimodal code
- Docs
- Python not yet supported