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
- [Bugfix] MambaManager.find_longest_cache_hit ignores drop_eagle_block, silent corruption with prefix caching + MTP/EAGLE
- [Bug]: Sparse NCCL corrupts non-contiguous patch indices and values
- [Bug] First dummy batch on a DP rank compiles MTP/EP Triton kernels at request time (wide-EP + MTP)
- [KVConnector][P2P] Register peers off the scheduler thread
- [Misc] Removing unreachable condition
- [Frontend][Responses] Reject namespace tool names that exceed the input budget
- [MRV1] Keep dummy warmups out of EPLB load stats
- fix(model): support multi-image Nemotron inputs
- [Bug]: DeepSeek-V4.1 produces NaN with CUDA graphs on SM120/121 — dummy capture batches poison the null KV block
- [Security]: remote media URLs allow SSRF to internal/link-local addresses by default
- Docs
- Python not yet supported