Ming Image 0.1 Design Layer is inclusionAI’s image-to-image model for separating a flattened design into RGBA layers. Select the exact ID inclusionai/ming-image-0.1-design-layer with the Images method at POST /v1/images/generations. Its role is to take an existing composition and a written layer plan, then return separate image outputs for the layers.
Supply exactly one reference image through input_references and describe what should be isolated in prompt. The documented reference representation accepts an HTTP(S) image URL or a base64 data URL. A useful layer plan can distinguish the background, the main object and foreground decoration, explaining which visible elements belong together. Use the actual flattened design as the reference so the decomposition is tied to the composition you intend to edit.
The dedicated model profile sets input_references to one reference and n to 1. That request setting describes one requested decomposition; the documented model behaviour returns one image per layer. Preserve every image in the response rather than treating the first data item as the whole result. The prompt should communicate the intended grouping of elements, while the returned layers provide the actual output set to inspect.
The supported output_format choices are png and webp. Output dimensions follow the input image. Explicit size and aspect_ratio values are rejected for this edition, so prepare the reference image at the dimensions required for the composition and leave those controls unspecified. The exact selector identifies the Design Layer edition, whose input is an existing design and whose output consists of image layers.
Generation uses a complete image response and this model’s dedicated profile does not support streaming. Decode each returned b64_json image or retrieve each returned image URL, keeping the actual output format when saving the file. Keep the layers together with the source image and layer plan. In a layer-aware editor, check the transparent regions, the separation of neighbouring elements and the reconstructed overall composition before making local changes.
The result is a set of raster image layers. It can support rearrangement, replacement of individual visual elements or further compositing in an editor, with the retained source available for comparison. Inspect any lettering and object edges that cross layer boundaries so the chosen decomposition remains useful for the intended revision.
", "aboutText": "Ming Image 0.1 Design Layer is inclusionAI’s image-to-image model for separating a flattened design into RGBA layers. Select the exact ID inclusionai/ming-image-0.1-design-layer with the Images method at POST /v1/images/generations. Its role is to take an existing composition and a written layer plan, then return separate image outputs for the layers.\n\nSupply exactly one reference image through input_references and describe what should be isolated in prompt. The documented reference representation accepts an HTTP(S) image URL or a base64 data URL. A useful layer plan can distinguish the background, the main object and foreground decoration, explaining which visible elements belong together. Use the actual flattened design as the reference so the decomposition is tied to the composition you intend to edit.\n\nThe dedicated model profile sets input_references to one reference and n to 1. That request setting describes one requested decomposition; the documented model behaviour returns one image per layer. Preserve every image in the response rather than treating the first data item as the whole result. The prompt should communicate the intended grouping of elements, while the returned layers provide the actual output set to inspect.\n\nThe supported output_format choices are png and webp. Output dimensions follow the input image. Explicit size and aspect_ratio values are rejected for this edition, so prepare the reference image at the dimensions required for the composition and leave those controls unspecified. The exact selector identifies the Design Layer edition, whose input is an existing design and whose output consists of image layers.\n\nGeneration uses a complete image response and this model’s dedicated profile does not support streaming. Decode each returned b64_json image or retrieve each returned image URL, keeping the actual output format when saving the file. Keep the layers together with the source image and layer plan. In a layer-aware editor, check the transparent regions, the separation of neighbouring elements and the reconstructed overall composition before making local changes.\n\nThe result is a set of raster image layers. It can support rearrangement, replacement of individual visual elements or further compositing in an editor, with the retained source available for comparison. Inspect any lettering and object edges that cross layer boundaries so the chosen decomposition remains useful for the intended revision.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "1812c6a7d28fc2f87d790ae302eb921ca421f63f7a575785b9e3363ac248ef26", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "inclusionai/ming-image-0.1-design-layerMing Image 0.1 Design Layer decomposes one flattened design image into separate RGBA image layers. Supply exactly one reference and a prompt describing the intended layer plan. This edition returns image layers for further composition, supports PNG or WebP, and preserves the input image’s dimensions.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Ming Image 0.1 Design Layer is inclusionAI’s image-to-image model for separating a flattened design into RGBA layers. Select the exact ID inclusionai/ming-image-0.1-design-layer with the Images method at POST /v1/images/generations. Its role is to take an existing composition and a written layer plan, then return separate image outputs for the layers.
Supply exactly one reference image through input_references and describe what should be isolated in prompt. The documented reference representation accepts an HTTP(S) image URL or a base64 data URL. A useful layer plan can distinguish the background, the main object and foreground decoration, explaining which visible elements belong together. Use the actual flattened design as the reference so the decomposition is tied to the composition you intend to edit.
The dedicated model profile sets input_references to one reference and n to 1. That request setting describes one requested decomposition; the documented model behaviour returns one image per layer. Preserve every image in the response rather than treating the first data item as the whole result. The prompt should communicate the intended grouping of elements, while the returned layers provide the actual output set to inspect.
The supported output_format choices are png and webp. Output dimensions follow the input image. Explicit size and aspect_ratio values are rejected for this edition, so prepare the reference image at the dimensions required for the composition and leave those controls unspecified. The exact selector identifies the Design Layer edition, whose input is an existing design and whose output consists of image layers.
Generation uses a complete image response and this model’s dedicated profile does not support streaming. Decode each returned b64_json image or retrieve each returned image URL, keeping the actual output format when saving the file. Keep the layers together with the source image and layer plan. In a layer-aware editor, check the transparent regions, the separation of neighbouring elements and the reconstructed overall composition before making local changes.
The result is a set of raster image layers. It can support rearrangement, replacement of individual visual elements or further compositing in an editor, with the retained source available for comparison. Inspect any lettering and object edges that cross layer boundaries so the chosen decomposition remains useful for the intended revision.
POST /v1/images/generationstext, imageimage| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits inclusionai/ming-image-0.1-design-layer | Default— |
Fieldprompt | Typestring | RequirementRequired | Values & limits | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits false | Defaultfalse |
Fieldinput_references | Typearray | RequirementRequired | Values & limits | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Maximum: 1 | Default— |
Fieldoutput_format | Typestring | RequirementOptional | Values & limits png, webp | Default— |
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, prompt, input_references |
Parameteroptional | Contractstream, n, output_format |
Parametermaterial_types | Contracttext, image |
Parameterconditions | ContractUse stream:false or omit stream. The complete buffered JSON result contains all generated images. |
Lifecycle: synchronous_result_then_owned_library
GET /v1/media/assets
GET /v1/media/assets/{asset_id}
The displayed rates estimate the cost before generation. The final charge is calculated from the cost recorded for the original generation. Measured output duration and forecast quantities are not the final billable quantity.
Tariff references do not add request options. Supported parameters are listed for the selected method.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Forecast quantity | Charge |
|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Forecast quantity— | Charge Input · Input tokens · Estimate for comparison only — / 1K tokens Output · Output tokens · Estimate for comparison only — / 1K tokens Input · Cache read · Estimate for comparison only — / 1K tokens Input · Cache write · Estimate for comparison only — / 1K tokens Input · Cache write, 1 hour · Estimate for comparison only — / 1K tokens |
| ConfigurationEstimated generation cost | Rate schedule— | Forecast quantityone generation estimate | Charge Input · Estimated generation cost · Estimate for comparison only $0.98304 estimate before generation This estimate is not an additional charge or a fixed rate for the completed generation. |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\nexport AILOOT_API_KEY\n: "${AILOOT_IMAGE_URL:?Set an actual customer material or documented value}"\nexport AILOOT_IMAGE_URL\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "inclusionai/ming-image-0.1-design-layer", "prompt": "A calm studio product photograph with a plain background.", "input_references": [{"type": "image_url", "image_url": {"url": "YOUR_ACCESSIBLE_IMAGE_URL"}}]}')\np['input_references'][0]['image_url']['url']=os.environ['AILOOT_IMAGE_URL']\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/images/generations \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\n# Submit once. Retain the complete result, original identifiers and every response header.\nDownload each returned image or save its Base64 contents
GET /v1/media/assetsGET /v1/media/assets/{asset_id}Save every completed image; retrieve each returned URL or decode its Base64 contents.
Retain every returned image and field. Use owned content_url values and actual asset_id records; refresh the same asset when a link expires.
Preserve the original response, headers and X-Client-Request-ID. An uncertain outcome does not authorize another creation.
Ailoot Ultra inclusionai/ming-image-0.1-design-layer
Use the method paths shown above.