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 supported21 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: Add environment variable support for configuring NIXL disaggregation backend
- [RFC]: Deprecate and delete InductorAdaptor in favor of InductorStandaloneAdaptor
- [EPLB]: Optimize export_load_view update
- Add **kwargs parameter to v1 FlashAttentionImpl as catch-all
- [Core] Add determinism warmup automation for batch invariant mode
- Fix #26037: Skip CUDA platform detection when displaying help
- [RFC]: All Ops should be determined during init and wrapped in a Layer Module to avoid envs.ENVIRON overhead
- [Bug]: Usage of VLLM_ALLOW_LONG_MAX_MODEL_LEN=1 in V1 likely to cause a crash
- raise 400 Bad Request with detailed error message for `aiohttp.ClientError`
- [UX] Move FlashInfer workspace cache inside of vLLM's cache (`~/.cache/vllm`)
- Docs
- Python not yet supported