ComfyUI v0.45.0 doubles Qwen3.x local text generation speed
AIComfyUI says its v0.45.0 update roughly doubles Qwen3.x token generation speed on the same hardware and checkpoints, without changing model quality. On an RTX 5090, Qwen3.5 9B rose from 106 to 219 tokens per second, and Qwen3.8 27B describing a 1536×1536 image rose from 55 to 122.
AIWhy it matters.
The post explains how ComfyUI's kernel, MTP, and attention changes raise Qwen3.x token rates, useful for judging local inference performance tradeoffs.
Source details
Comfy Blog · rattus128 · blog.comfy.orgPublished · added here
Original title: Doubling LLM token rates in ComfyUI