Diffthinker Multimodal Reasoning

Apply diffusion models as native generative agents for vision-centric reasoning tasks (sequential planning, constraint satisfaction, spatial configuration) instead of text-based LLM chains. Achieves 3x+ improvements over GPT-5 and Gemini-3 on visual reasoning. Use when image-to-image generation better captures the reasoning constraints than text-based problem decomposition.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 3bef743b84

Frequently asked questions

npx get-skill add adu2021/diffthinker-multimodal-reasoning