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
- [KVConnector] Add priority-aware transfer ordering
- [Bugfix][Reasoning] Say why thinking_token_budget is unavailable
- [Bugfix][RL] Fail a weight update without taking the engine down
- [Feature][KVConnector][DSv4.1] Encoder-only prefill with Mooncake P/D
- [CPU][Zen] Prepack quantized weights and add DA8W4 for zentorch on Zen CPUs
- [Bugfix][KV Connector] Dedupe DecodeBenchConnector KV fills
- [ROCm][Bugfix] Make the RoPE + KV-cache fusion reachable by explicit request and through its documented opt-ins
- [Bug]: `/v1/chat/completions/batch` leaks hidden reasoning through logprobs and token IDs when `include_reasoning=false`
- [Bug]: AssertionError in _update_from_kv_xfer_finished when LMCache KV load fails on heterogeneous attention models
- [Bug]: Spec decode hits illegal memory access when the draft model's context is shorter than the target's
- Docs
- Python not yet supported