> Update (2026-09-28): I trained this on a bunch of edit pairs to reinforce pixel consistency, > and it turns out the dataset might have overfit it on certain types of edits. I've heard it > ignores movement, like a new pose or a turned head. I plan to try a v2 once I get the dataset > put together. Until then you can try out the two workflows I uploaded: > > - `qwen-image-2.1-edit-consistency
Qwen Image 2.1 edits don't stay where you put them. Ask for a watercolor, a comic or an anime version and the picture comes back a few percent taller, sometimes nudged sideways, different for every seed: a waistband 40 px lower, a face that no longer lines up with the original. On a local edit it also repaints things you didn't ask about (hair strands, signs, texture). This LoRA makes the edit **o
Highlights
- ostris/ai-toolkit, arch: qwenimage2, Comfy-Org INT8
- 950 edit pairs from 257 pictures: 110 portraits we rendered with Qwen Image 2.1 for this,
- Each pair started as a real Qwen Image 2.1 edit of one of those pictures: restyles, relighting,
- forward pairs (picture -> edit): the picture is moved onto the edit's frame and the edit
- reverse pairs (edit -> picture, "turn this watercolor into a photo"): the edit is moved onto
- local edits keep the original's pixels everywhere outside the edited thing.
- Captions are the edit instructions themselves, about a third of them followed by a "keep the
- Rank 32 / alpha 32, AdamW8bit, LR 1e-4 constant, batch 1, shift timesteps, resolution: 1408
From the source
Images from ausboss on Hugging Face, shown with a link to the original.