| --- |
| 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. |
|
|
