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
- Add SparDA lookahead KV connector
- [Bug]: Stale partial prefix-cache hash resurrected into the cache after full-block promotion
- [Kernel] Avoid materializing FP32 rotary Q/K tensors
- [Bugfix] Handle empty or refusal content when reconstructing Responses input messages
- [Bugfix][KVConnector] Release executor and lock fds in ExampleHiddenS…
- [Bug]: Latency histograms get host-uptime-sized values when a request finishes before reaching a milestone
- [Bugfix] Avoid nonsense latency intervals for unreached milestones
- Enable MLA RoPE KV-cache fusion by default
- [Bug]: Minimax-M3 running crase
- [Bugfix][Memory] Load weights outside the CuMem pool and re-home them for sleep mode
- Docs
- Python not yet supported