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 supported49 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: Improve engine RPC failure handling and shutdown
- [Bugfix][MRV2] Reserve encoder graph memory without decoder graphs
- [Feature] [KV Offload] Support sharing SimpleCPUOffload prefix caches across local DP replicas
- [Bug]: An Anthropic messages test is not compatible with Anthropic SDK 1.x
- [WIP] Support fused checkpoint loading in transformers backend
- [RFC]: Increase vLLM Review Capacity with a Reviewer Role
- [Performance][Hardware][RISC-V] Use a 16-column RVV tile for narrow-M attention GEMMs
- [Kernel][ROCm] change prefix_prefill kenrel num_warps=8 for RDNA GPUs
- [Bugfix][MLA] Fix SM90 sparse MLA IMA on >16k-row steps by chunking plan/run
- [ROCm][MoE] Support MiMo-V2.6 MXFP4 on gfx942
- Docs
- Python not yet supported