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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Core] KV cache: block-range eviction for live requests
- [Build] Skip phantom outputs in strict editable installs
- [Core] Bounded-memory streaming sessions: retention, in-place KV eviction, consolidation
- [Frontend] Streaming session API (/v1/streaming)
- [Quantization] Support Humming W4A8 Block in CT
- [Bugfix] Resolve callable hf_overrides in get_quant_config
- [Bugfix][KV Connector] Avoid blocking on a full event queue
- [Bug]: MTP draft allocates a second full vocab embedding, OOMing a 27B INT4 target that the baseline fits
- [KV offload] Scale uniform-type attention groups by DCP in build_offloading_config
- [Installation]: 0.28.0 build error in jetson agx orin 64G jetpack 7.2.1
- Docs
- Python not yet supported