Qwen-Image Text-to-Image LoRA Distillation Accelerated Version Workflow / Download

https://www.liblib.art/modelinfo/274af66173124d4eaf7864288036e6e0

About Qwen-Image LoRA Distillation Acceleration Model

Model Description: This model is a distillation accelerated LoRA for Qwen-Image. We followed the training pipeline of the model DiffSynth-Studio/Qwen-Image-Distill-Full, converting trainable model parameters to LoRA, making it easier to integrate into various image generation frameworks.

The training framework is built on DiffSynth-Studio. The training data consists of 16,000 images generated by the original model using prompts randomly sampled from DiffusionDB. The training program ran on 8 * MI308X GPUs for about 1 day.

* Training of this model received technical support from the ModelScope Community

* Provided by DiffSynth-Studio

Reprinted from: Universal Qwen-Image LoRA Distillation Acceleration Model - Generation Speedup & Image Quality Enhancement