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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Handle OOV multimodal prompt logprobs
- [Performance]: Avoid extra logits passes for temperature-only sampling
- [Bugfix][EPLB] Fix startup failure with MoE layers on multiple pipeline stages
- [Bugfix] Stop advertising fp8_e5m2 where the KV writer stores e4m3
- [Bugfix] Partition HF3FS external cache keys by multimodal and LoRA identity
- [Doc] Raise the bar for in-tree kernel translation units
- [Feature]: Carry the HTTP EPD multimodal handoff on the Rust gRPC frontend
- [Doc] Correct docstrings that promise a return value
- [Kernel][MiniMax-M3] Drop tl.make_block_ptr from the Hopper sparse prefill
- [Spec Decode] Broaden NVFP4 draft LM-head validation beyond GB10: Memra results and Blackwell/model matrix
- Docs
- Python not yet supported