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 supported26 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: `/v2/embed` with `input_type` returns misleading 400 on nemotron-embed-vl
- [vllm IR]: Remove `QuantFP8` in favour of direct `ir.ops` calls
- [Bug]: Qwen/Qwen3.6-27B - Batch Inference still not working in vllm 0.19.1
- [Bug]: Gemma4 0% prefix cache hits with hybrid attention + DFlash
- [WIP][kv_offload] Decouple store policy and request lifecycle from the scheduler
- [Feature]: Integrate fused `kMoEFinalizeARResidualRMSNorm` from FlashInfer
- [Kernel] Give a helpful error when benchmark_moe.py hits an unknown arch
- [Feature]: Support responses api in tokenize endpoint
- [Bug]: UX of Weight Prefetchers on Startup Failure
- [RFC]: Fault tolerant EPLB
- Docs
- Python not yet supported