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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix][ROCm] Make Mooncake an optional connector dependency
- [Kernel][MoE] Add small-batch BF16 SiLU Triton expert path on Hopper
- [Docs][Spec Decode] Qualify Gumbel recovery watermark strength
- [Bugfix][Frontend] Keep special tokens when truncating from an explicit side
- [Bugfix][Spec Decode] Log EAGLE draft RoPE parameters at startup
- [ROCm][Misc] Write GDN prefill output in place for the Triton and FlyDSL backends
- [Model] Support ModernBERT decoder (Ettin) via the Transformers modeling backend
- [Core] Shared abstractions for specialized decode kernels
- [Perf][DSv4.1] Fuse C1/C2 compression with FP8 KV cache insertion
- [ROCm][CI][BugFix][AITER ON] Make LM Eval Small Models tests pass on MI300/MI355 and ensure they can work with vllm serve and vllm bench serve with AITER on and off
- Docs
- Python not yet supported