Noob tip: pull the exact ComfyUI workflow back out of your old images
So I’ve only been messing around with Krea2 in ComfyUI for a little while, and I ran into the classic problem: I make an image I really like, then a week later I want the exact same settings back and have no clue what I did.
Proof1 community thread
Key points
- Ended up asking Claude to build me a prompt and workflow extractor that pulls the prompt and workflow out of my generated pictures.
- Now I just drop an image in, it spits out a .json, and I drag that into ComfyUI.
- Always looking for more quality of life stuff like this, please feel free to tell any QoL regarding generation/comfyui!
- EDIT: omg thank ya'll for mentioning the just dropping the image into comfyui! makes my life even easier haha
Sources (1)
- [1]Noob tip: pull the exact ComfyUI workflow back out of your old imagesr/StableDiffusion (top, day) · Oct 13, 07:40 PM
So I’ve only been messing around with Krea2 in ComfyUI for a little while, and I ran into the classic problem: I make an image I really like, then a week later I want the exact same settings back and have no clue what I did.
Ended up asking Claude to build me a prompt and workflow extractor that pulls the prompt and workflow out of my generated pictures.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 13, 2026Run fast, private agents on-device with LFM2.5-24B-A2B
- Oct 13, 2026Did a quick anime-to-realism test in MiniMax H3 with just a video reference + lora
- Oct 13, 2026inclusionAI/Ling-3.1-flash
- Oct 13, 2026Portable .char consistent characters(same face, body, cloths & voice) - MinimaxH3, Flux2 & Krea2
- Oct 13, 2026Free Krea2 Character & Style Catalog v0.9 — 3,919 samples, reusable prompts, and ComfyUI workflows
- Oct 13, 2026Hybrid Reasoning Agents
