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
- [Bugfix] PCP: replicate prefills too short for every rank, also without DCP
- [Frontend] generate: keep sample logprobs as engine rows and render them without objects
- [PoC] Transfer hidden states for P/D
- [Bug]: stream_interval is ignored for RequestOutputKind.CUMULATIVE (the SamplingParams default)
- [Bugfix][Frontend] Honor stream_interval for CUMULATIVE request outputs
- [Bugfix][Frontend] Clean up failed online beam searches
- External speculators (DSpark Qwen3 draft, DFlash2) crash with kv_cache_dtype=nvfp4_ds_mla on GLM-5.3 NVFP4: draft-model code assumes split k/v
- Handle TorchCodec native library load failures
- [Frontend] mv run_rust_frontend to launchers.
- [Core] Make the V2 sampler allowed_token_ids and logit_bias caps env-tunable
- Docs
- Python not yet supported