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 supported3 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Installation]: Is there no more "***.whl" based on cuda12?
- [Bug]: Poor TTFT performance with simultaneous --enable-chunked-prefill and --enable-prefix-caching
- [Usage]: FP8 and INT8
- [Bug]: sm75 --num-scheduler-steps 8, unhandled errors in a TaskGroup
- [Bug]: In the case of quantization=compressed-tensors, Qwen2-57B-A14B-Instruct is not supported.
- [Misc]: kvcache hash collision
- [Misc]: Throughput calculation in benchmark_throughput.py
- [Bug]: requests with response_format cause vllm to hang with pipeline parallel
- [Usage]: How to determine the batch size for batch offline inference?
- Reshape cache flash kernel to support HND layout
- Docs
- Python not yet supported