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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: NIXL handshake crashes with StopIteration for Mamba-only models (no FullAttention KV group)
- [Mamba] Declare prefill checkpoint alignment once per layer
- [MRV2][DBO] Don't microbatch a step that splits a prefix from its writer
- [Bugfix][Kimi-Linear] Honor --mamba-ssm-cache-dtype for the KDA recurrent state
- [Bugfix][KV Connector] Replace HF3FS block table on MRV1 resume
- [Bugfix][KV Connector] Drain and validate HF3FS save writers
- [Bug]: /v1/systemone tokenizes the whole state once per question, so decision latency grows with questions × state length
- [Bugfix][Kimi-K3] Keep speculative scratch out of checkpoint history
- [Perf][torch.compile] Pre-create FlashInfer AR+RMSNorm workspaces at backend init
- [Security] Support stream_options.include_obfuscation to mitigate token-length side channels
- Docs
- Python not yet supported