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 supported31 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [KV Offload] Async CPU memory pinning for SimpleCPUOffloadConnector
- [Bugfix] NIXL PD: Don't transfer spec-decode lookahead blocks
- Fix(rope): preserve rope_theta in rope_parameters when loading transformers v5 checkpoints with transformers v4
- [Core] feat: Widen Request.priority from int to float
- [Model] Add HyperCLOVAX-SEED-Think-14B reasoning and tool parsers
- [Frontend] Add HTTP/2 support via Hypercorn (--enable-http2)
- [Perf] Add tuned Triton fused MoE configs for NVIDIA H20
- [Bugfix][Benchmark] Fix dataset stats plot filename
- [CPU][Zen] Route BF16 MoE inference through zentorch on AMD
- [Config] Add field validators for attention, KV/EC transfer, and KV events configs
- Docs
- Python not yet supported