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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Fix MiniMax M3 index_topk kernel for non-power-of-2 num_idx_heads (#49157)
- [Bug]: Engine core livelock (100% CPU, no crash) with MTP speculative decoding + xgrammar structured outputs - regression from v0.24.0
- [Perf] Shape-adaptive tile config for batch-invariant matmul on SM80
- [Bug]: StreamingParserEngine leaks a bare tag fragment (e.g. '<tool_') into content when a PreLexedTerminal supersedes a buffered text-token prefix of the same tag
- [Model] Honor fp32 head_dtype in Inkling muP logits (target + MTP draft)
- [Bug]: Native KV offloading with prefix caching changes tool visibility after a long shared-prefix request on Qwen3.6
- [Bugfix][Quantization] Respect explicit WNA16 MoE backend
- [Bug]: auto_round (INC) path hardcodes lm_head_quantized=False, crashing on checkpoints with quantized lm_head that auto-round legally produces
- TRITON_ATTN support for KV cache dtype fp8 on SM75 to pre-SM89
- fix(security): validate allowed_token_ids against model output vocab size
- Docs
- Python not yet supported