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 supported13 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: When compiling with ranges, we should pass the range information to Inductor
- [Bug]: [H200] DeepSeek R1 `--enable-expert-parallel True` is slower than `False`.
- [Bug]:
- [Bugfix] fixes shared expert forwarding in fused_moe.
- [amd] enable kimi k2 fp8 kv on amd mi300
- Add SamplingParams.shallow_copy to only clone mutable fields
- [Bug]: Custom logits processor edge-cases
- Fix Dynamo trace errors for CPU offloading without UVA.
- [Performance]: Fully Async Spec-Decoding | Make `seq_lens_cpu` in CommonAttentionMetadata optional
- [Bug]: cpu-offload-gb crashes with "CPU tensor must be pinned"
- Docs
- Python not yet supported