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
- [DSv4.1][PCP] Share CPU-offloaded Engram tables
- fix(benchmarks): harden SSE handling
- [Bugfix][XPU] Skip Triton allocator registration for TD scaled_mm on XPU
- [Bugfix] Guard FlashInfer seq_lens_cpu_upper_bound shortcut with async_scheduling
- [Bugfix][NIXL] Fix handshake crash for SSM-only models
- [ROCm][AMD][AITER][DiffusionGemma]: RedHatAI/diffusiongemma-26B-A4B-it-FP8-dynamic performance slower on MI300/MI355 with AITER on vs AITER off
- [Bug]: `reset_prefix_cache(reset_running_requests=True)` during an in-flight prefill loses the request's prompt logprobs
- [XPU] Add Qwen4Exp (Qwen3.8-Flash-Next) model architecture support
- [Bug]: NIXL handshake crashes with StopIteration for Mamba-only models (no FullAttention KV group)
- [Mamba] Declare prefill checkpoint alignment once per layer
- Docs
- Python not yet supported