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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [MoE][DeepEPV2] Add PaddedStandard activation format
- [Tool Parser] Add an Inkling structural tag so forced tool choice is enforced
- [Bug]: ModelOpt FP8_PB_WO checkpoints using the weight_scale_inv / rank-2 export convention fail to load
- [WIP]Support PCP with DP
- [Feature]: MORE VELOCITY TO INFERENCE
- [RFC]: Backend-stable stateless RNG for Gumbel sampling
- [Rust Frontend] Add pooling support to engine core client and LLM
- [Bugfix] Bound client-supplied max_tokens on /inference/v1/generate
- [Bug]: FlashMLA sparse prefill assertion failed on H20 with DeepSeek-V4-Flash + DSpark + long context (v0.26.0 regression)
- Use `dtype` instead of deprecated `torch_dtype` for transformers >= 4.56
- Docs
- Python not yet supported