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 supported36 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: CPU offloading fails with block_size=256 + speculative decoding due to eagle group block_size mismatch
- Fix Mooncake sender listener startup with IPv6 hosts
- [Bug]: Qwen3.5 CausalLM missing embedding_modules breaks LoRA on embed_tokens/lm_head
- [Spec Decode] Context-length-aware K in DSD (RFC #48627): extend num_speculative_tokens_per_batch_size with a ctx axis
- [Core] Add opt-in eager PyNccl TP all-reduce split for PIECEWISE CUDA graphs
- [Bug]: NVFP4 garbage output under torch.compile when per-rank K is not 32-aligned (TP8 on SM120): padded block-scale column reads stale memory; eager/Marlin/aligned-TP correct
- Skill to check for prebuilt wheels during installation
- [Bugfix](sampling): search Qrita top-p pivots in log space
- [Core] Add remaining copy counts to KV removal events
- [Bugfix] Fall back to native sampling when FlashInfer cannot target the GPU
- Docs
- Python not yet supported