README.md

---
tags:
- diffusion-single-file
base_model:
- inclusionAI/Ming-Image-0.1-Design
---

You will find here all the Ming-Image-0.1-Design models used with WanGP (https://github.com/deepbeepmeep/Wan2GP) :


WanGP by DeepBeepMeep : The best Open Source Video Generative Models Accessible to the GPU Poor

WanGP supports the Wan (and derived models), Hunyuan Video, Minimax H3, Krea-2, Flux 1 & 2, Qwen Image 1/2.1, Z-Image and LTX-2, LTX Video models with:

Low VRAM requirements (as low as 6 GB of VRAM is sufficient for certain models)
Support for old GPUs (RTX 10XX, 20xx, ...)
Very Fast on the latest GPUs
Easy to use Full Web based interface
Auto download of the required model adapted to your specific architecture
Tools integrated to facilitate Video Generation : Mask Editor, Prompt Enhancer, Temporal and Spatial Generation
Loras Support to customize each model
Queuing system : make your shopping list of videos to generate and come back later
Discord Server to get Help from Other Users and show your Best Videos: https://discord.gg/g7efUW9jGV

Follow DeepBeepMeep on Twitter/X to get the Latest News: https://x.com/deepbeepmeep

## Ming Image checkpoints

WanGP includes Ming Image 0.1 Design for image generation and editing, and Design-Layer for decomposing a reference image into RGBA layers. Both use the same Bailing language model and vision tower. The Bailing core is in `BailingMM2-Ming-Image/` alongside its tokenizer, and the vision tower plus image projection are in `ming_image_shared/`. Each variant's connector, FFN and conditioning projections are in `ming_image/conditioning_*` or `ming_image_layer/conditioning_*`. Select matching BF16 or INT8 ConvRot files for all three encoder pieces.

The diffusion transformer single-file checkpoints remain at the repository root. Design and Design-Layer use their respective transformer and conditioning weights and share the VAE in `ming_image/`. WanGP downloads and loads the required pieces automatically. The vision tower is a separate MMGP model, so text-only generation does not need to load it to the GPU.

Upstream checkpoints: [Design](https://huggingface.co/inclusionAI/Ming-Image-0.1-Design) and [Design-Layer](https://huggingface.co/inclusionAI/Ming-Image-0.1-Design-Layer). The released code and model assets are MIT licensed.