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][Core] Count the external claim's state block in MambaManager admission
- [Docs] Document NIXL KV connector metrics aggregation semantics
- [RFC] RL Weight Update Roadmap
- [Feature]: [RFC] Dynamic Latent-Trajectory Consensus Governor for Reasoning Models (DeepSeek-R1 / QwQ) to Free KV-Cache 4.5x Faster
- [Bug]: Intermittent Xid 31 MMU fault in pynccl all_reduce during CUDA graph
- [Bug]: MRotaryEmbedding + YaRN uses 4x original_max_position_embeddings for the YaRN correction range and cache size
- [Feature]: Allow LBNHC/NHD for ROCM_AITER_UNIFIED_ATTN where supported
- [Bugfix][Quantization] Name the layer when a MoE expert weight is unquantized
- [Perf][Core] Restore compact prefix-cache entries after duplicate eviction
- [Bugfix] Keep FlashAttention sparse ahead of FlashInfer SM90 for NoPE-512 MLA
- Docs
- Python not yet supported