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 supported30 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Core] Add prefix-cache-aware routing for internal DP load balancing
- Feat/unified comm abstraction
- [RFC] NSA is architecturally incompatible with NVIDIA consumer/workstation GPUs — MLA is the only viable sparse attention path
- [Usage]: Support Gemma 4 E4B, 31B, 26B-A4B, and assistant variants (MTP) on TPU v6e 1x1 with vLLM
- Fix MRV2 Gumbel sampling for non-finite logits
- [Bug]: `vllm serve --config=...` accepts the config path but does not expand the YAML
- fix: prevent silent performance degradation by allowing custom all-reduce for cap-equal tensors
- [Perf][Core] Use power-of-two-choices for DP engine selection at large DP
- [Bug] FusedMoE `_load_per_tensor_weight_scale` rejects shape `(1,)` per-tensor scales
- [Hybrid SSM] Investigate accuracy divergence between `mamba_chunk_scan` and `selective_state_update` kernels
- Docs
- Python not yet supported