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
- [Benchmark] Add skewed (zipf) expert-load coverage to benchmark_moe.py
- [RFC]: Programmable KV Cache: Composable Policies for Agentic Serving
- [RFC]: [FS Offloading][ThreadPool] Updates to Thread Pool
- [Bugfix][ModelLoader] Make enable_weights_track check quantized layers
- [Perf][DSA] Add FlashInfer gvr_2 as a sparse-indexer decode top-k backend
- [RFC]: Checkpoint-aware cache eviction and segmented recomputation for hybrid models
- [Kernel][PCP] Publish direct-final KV into ExtensibleKVCache storage
- Fix Qwen vision encoders falling back to data parallelism on non-divisible TP sizes
- [Kernel] Keep eps in fp32 in the batch-invariant RMSNorm Triton kernels so torch.compile matches eager
- [Bugfix][ROCm] Guard compressed-tensors FP8 KV cache auto-selection
- Docs
- Python not yet supported