Gradio open-sources doodle-in LoRA for Qwen-Image 2.1 with full training data
Gradio · x · 2026-09-29
Gradio released Qwen-Image-2.1-doodle-in-LoRA on Hugging Face: a scribble-to-object edit LoRA for Qwen-Image 2.1 that lets you draw an outline on a photo, name the object, and watch it appear in the photo's own art style.
- Fully open: the adapter (diffusers and ComfyUI files), all 6,042 training pairs plus 160 test pairs, every eval output, and the build/training/eval scripts
- Works best with checkpoint 500 + Viggle turbo, finishing edits in 6 steps — image editing control that's hard to get from prompts alone
- Released under the Qwen Research License (non-commercial); training photos from Open Images (CC BY 2.0)
- A live ZeroGPU demo is available: draw a magenta outline, name the object, done
Related event: Gradio Open-Sources Doodle-in LoRA for Qwen-Image 2.1(5 posts)→
More from Multimodal
- Testing Qwen Image 2.1 and Minimax H3 on anime-to-photo conversion and film photo restoration — Ant_6431 · 2026-11-16
- Second take of a 'Cubee run' video generated with Minimax H3 — JahJedi · 2026-11-16
- Non-artist builds 14-chapter hand-painted storybook game with Fable 5.2 in 4 hours — jradoff · 2026-11-16
- Subject LoRAs on Minimax H3 Image learn poor likenesses while identical datasets excel on Qwen 2.1 — Ad_Organic · 2026-11-16
- Seedance-generated anime train scene stuns Reddit with cinematic quality — Ok-Solution-7889 · 2026-11-16
- AI-generated 'Men in Black' reimagined in 1977 goes viral — techhalla · 2026-11-16