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
- [chore] remove kernel_block_sizes duplicate check in get_preferred_block_size
- [Usage]: Unevenly allocate gpu memory
- [Bug]: ~11% long-prefill regression v0.24.0 -> v0.25.x on SM120, traced to the FlashInfer 0.6.13 bump (#46683)
- [Bug]: Prefix caching silently inert on hybrid Mamba2 models when max_model_len < auto-selected block size
- [Bugfix] Warn when prefix caching is inert on hybrid models (block_size > max_model_len)
- [Bug]: MiniCPM-V-4.6 fails to load in vLLM 0.25.0: qkv_proj is incorrectly mapped to qkqkv_proj
- fix(sampling_params): check top_k type before comparing it to -1
- [Bug]: Local PP Rank0 stalls in legacy non-SPMD Ray AsyncLLMEngine under imbalanced pipeline stages
- Revert "[Bugfix] Guard mixed-dtype allreduce RMSNorm quant fusions" (#48330)
- HybridKVCacheCoordinator: DSpark draft-model KV-cache group's ephemeral content incorrectly vetoes GPU-tier prefix cache reuse
- Docs
- Python not yet supported