Image modelOpen weights
Ming-Image-0.1-Design-Layer
InclusionAI
- Released
- -
- Data date
- October 3, 2026
Ming-Image-0.1-Design-Layer takes a flattened design and a layer plan, then decomposes it into requested RGBA layers, unlike Ming-Image-0.1-Design. The recommended working resolution is 1024, or 512 for faster runs.
Its separate MIT checkpoint saves outputs as individual RGBA PNG files while preserving the input aspect ratio. Ming-Image-0.1-Design-Layer is therefore a decomposition model, not a second image-generation variant.
Specifications and access
| Specification | Value and source |
|---|---|
| Model ID | inclusionAI/Ming-Image-0.1-Design-LayerSource |
| Model class | Image-to-image model for layer decompositionSource |
| Input | Image and layer plan or requested layer countSource |
| Output | Separate RGBA PNG files per layerSource |
| Working resolution | 1024 recommended, 512 for faster runs, aspect ratio is preservedSource |
| Hardware | One CUDA GPU with 80 GiB VRAM in the validated configurationSource |
| License | MITSource |
Sources and data date
Every statement links to its underlying documentation or leaderboard.
| Type | Evidence and data date |
|---|---|
| Research status | Research date October 4, 2026. 1 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not found in the inspected sources | quality-benchmark Original wording The Crello table labels neighbouring methods, not Ming-Image-0.1-Design-Layer. This image-to-image layer model has no exact quality row. |
| Additional source | Ming-Image repository (retrieved October 3, 2026) · Ming-Image repository · Editorial description reviewed October 3, 2026 |
| Additional source | Ming-Image Design Layer model card (retrieved October 3, 2026) · Ming-Image Design Layer model card · Editorial description reviewed October 3, 2026 |