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
- [Bug]: Hidden-state extraction inflates KV cache memory requirements by 24× on a hybrid Mamba/MLA model
- [Rust Frontend] Accept native sampling parameters over gRPC
- [Bug]: Native KV offload advances next_stored_block_idx for every KV group when no keys were stored
- [Bugfix] Bill hidden-state extraction at its own KV cache group width
- [Bug]: --moe-backend marlin is applied to the unquantized MTP draft MoE and aborts (Qwen3.8-Flash-Next NVFP4 + speculative mtp)
- [Bugfix][Tool Parser] Make pythonic tool parsers agree across streaming modes
- [Metrics][KV Offload] Add vllm:kv_offload_config_info - generic info metrics for offloading the offloading connector
- [RFC]: Plain-Python watermarking performance regression benchmark
- [Bug]: race in `hadacore_transform`'s tail warp, last rows % 8 rows read stale shared memory when rows % 8 is 1-4
- [Model][Engram] Add Mooncake-backed DeepSeek V4.1 tables
- Docs
- Python not yet supported