Skip to content
ComfyUI Wiki · 工作流与开源创作分享·· 17 days agoAI score55

Qwen-Image 2.1 Lightning v2 RL released, four-step sampling full transformer model

Text to image

AI briefing

LightX2V team releases Qwen-Image 2.1 Lightning v2 RL, upgraded from v1's 67MB LoRA to a full transformer (28.5GB FP32), refined via RL to support four-step sampling, covering text-to-image, single-reference editing, and multi-reference generation.

Open original comfyui-wiki.com