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
- [Loader] Pinned host staging for the Run:ai streamer path
- [Reload] Keep captured-graph tensors valid across reload_weights after level-2 sleep (NVFP4 MoE + MLA)
- [Bug]: jit_monitor_mode="error" kills the engine on the first real request for hybrid mamba models — JIT warmup never covers precopy_mamba_align_fused_kernel
- Log EAGLE draft rope base at startup; warn on silent default
- [BugFix][Spec Decode] Log EAGLE draft RoPE and warn when config omits rope keys
- [BugFix] Combine return_sampling_mask SamplingParams precondition errors
- [BugFix][Quantization] Surface FP8 kernel rejection reasons for fp8_per_channel
- [Doc]: Advisory GHSA-mm38-5g96-7j6x lists affected versions as <= 0.8.5, but the code it names ships in 0.10.2 to 0.24.x
- [Perf][Frontend] Add bounded segment tokenizer cache to HF renderers
- [Kernel][Spec Decode] Fuse temperature scaling into the rejection-sampling kernels
- Docs
- Python not yet supported