AI Portrait Generator · MiniMax

Create an AI portrait or scene from a short description. Choose square, wide, or tall; add a link to one portrait photo only if you want to guide a person's appearance.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Create one still image from the user's description. Use a square shape unless they ask for a wide or tall image. If they provide one public link that opens a front-facing JPG, JPEG, or PNG portrait under 10 MB without signing in, use it to guide the person's appearance; otherwise create the image from text alone. Do not promise an exact likeness. Check the current price and use only the budget the user approved. Return the actual image, remind them to save it within 24 hours, and reuse the saved result when showing it again instead of generating another image.

Task:
Create one image of [a small dog sitting beside a sunny window]. Use [square, wide, or tall] if I have a preferred shape. If I want to guide a person’s appearance, I can share [a link to one front-facing portrait photo]. Remind me to save the image within 24 hours.

Paste into your connected AI's conversation.

Connection guide

Up to 0.0378 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

MiniMax

AI Portrait Generator · MiniMax

Create an AI portrait or scene from a short description. Choose square, wide, or tall; add a link to one portrait photo only if you want to guide a person's appearance.

You receive

Create one still image from your description. Choose a square, wide, or tall shape, and optionally add one portrait photo link to guide a person's appearance.

Start with your task

Create one image of [a small dog sitting beside a sunny window]. Use [square, wide, or tall] if I have a preferred shape. If I want to guide a person’s appearance, I can share [a link to one front-facing portrait photo]. Remind me to save the image within 24 hours.

What to know

A reference photo must be a front-facing JPG, JPEG, or PNG portrait under 10 MB with a link that opens without signing in. The image may not match the person exactly. Save it within 24 hours, before its download link expires.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Image request branch`body`	`object`	Optional	Provide model and prompt for a usable request; image-01-live additionally needs subject_reference. The root schema does not formally require body, but an empty root object does not describe a generation task.
Requested image count`body.n`oneOf · 1	`integer`	—	Integer 1–9, default 1. Controls the requested image quantity and per-result charging; inspect the current credit ceiling and actual returned count. Default `1` Minimum `1` Maximum `9`
Random seed`body.seed`oneOf · 1	`integer`	—	Integer from -9007199254740991 to 9007199254740991 for requesting reproducible generation. Record other settings too; a seed does not promise identical output across changed settings or model versions. Minimum `-9007199254740991` Maximum `9007199254740991`
Image model`body.model`oneOf · 1	`string`	—	Required within either branch: image-01 for prompt-based generation with an optional portrait, or image-01-live with a required portrait reference. Fixed value `image-01`
Image-01 width in pixels`body.width`oneOf · 1	`integer`	—	Only image-01 accepts this integer width, 512–2048 pixels, paired with height and divisible by 8. An explicitly supplied aspect ratio takes priority. Minimum `512` Maximum `2048`
Image-01 height in pixels`body.height`oneOf · 1	`integer`	—	Only image-01 accepts this integer height, 512–2048 pixels, paired with width and divisible by 8. It is not an accepted image-01-live setting. Minimum `512` Maximum `2048`
Image prompt`body.prompt`oneOf · 1	`string`	—	Required nonempty text, 1–1500 characters, describing the desired image. It is not an image caption returned as verified output. Minimum length `1` Maximum length `1500`
Aspect ratio`body.aspect_ratio`oneOf · 1	`string`	—	1:1, 16:9, 4:3, 3:2, 2:3, 3:4, 9:16 or 21:9; default 1:1. When explicitly supplied with image-01 width/height, the ratio takes priority. Default `1:1` Options `1:1, 16:9, 4:3, 3:2, 2:3, 3:4, 9:16, 21:9`
Image delivery format`body.response_format`oneOf · 1	`string`	—	url or base64, default url. URL links expire after 24 hours; base64 carries inline encoded image data rather than an expiring link. Default `url` Options `url, base64`
Automatic prompt optimization`body.prompt_optimizer`oneOf · 1	`boolean`	—	Optional boolean, default false, enabling automatic prompt optimization. It does not guarantee image quality or exact adherence to the original prompt. Default `false`
One portrait reference`body.subject_reference`oneOf · 1	`array`	—	Exactly one portrait reference when supplied, guiding the character’s appearance. Required for image-01-live, optional for image-01; only type=character is supported. Minimum items `1` Maximum items `1`
[]`body.subject_reference[]`oneOf · 1	`object`	—	Required fields `type, image_file`
Portrait subject type`body.subject_reference[].type`oneOf · 1	`string`	—	Required in the reference item and fixed to character. It does not add support for a list of arbitrary subject types. Fixed value `character`
Portrait image file`body.subject_reference[].image_file`oneOf · 1	`string`	—	Required nonempty public image URL or base64 data URL. JPG/JPEG/PNG under 10 MB; a single front-facing portrait is recommended rather than a guaranteed identity match. Minimum length `1`
Requested image count`body.n`oneOf · 2	`integer`	—	Integer 1–9, default 1. Controls the requested image quantity and per-result charging; inspect the current credit ceiling and actual returned count. Default `1` Minimum `1` Maximum `9`
Random seed`body.seed`oneOf · 2	`integer`	—	Integer from -9007199254740991 to 9007199254740991 for requesting reproducible generation. Record other settings too; a seed does not promise identical output across changed settings or model versions. Minimum `-9007199254740991` Maximum `9007199254740991`
Image model`body.model`oneOf · 2	`string`	—	Required within either branch: image-01 for prompt-based generation with an optional portrait, or image-01-live with a required portrait reference. Fixed value `image-01-live`
Image prompt`body.prompt`oneOf · 2	`string`	—	Required nonempty text, 1–1500 characters, describing the desired image. It is not an image caption returned as verified output. Minimum length `1` Maximum length `1500`
Aspect ratio`body.aspect_ratio`oneOf · 2	`string`	—	1:1, 16:9, 4:3, 3:2, 2:3, 3:4, 9:16 or 21:9; default 1:1. When explicitly supplied with image-01 width/height, the ratio takes priority. Default `1:1` Options `1:1, 16:9, 4:3, 3:2, 2:3, 3:4, 9:16, 21:9`
Image delivery format`body.response_format`oneOf · 2	`string`	—	url or base64, default url. URL links expire after 24 hours; base64 carries inline encoded image data rather than an expiring link. Default `url` Options `url, base64`
Automatic prompt optimization`body.prompt_optimizer`oneOf · 2	`boolean`	—	Optional boolean, default false, enabling automatic prompt optimization. It does not guarantee image quality or exact adherence to the original prompt. Default `false`
One portrait reference`body.subject_reference`oneOf · 2	`array`	—	Exactly one portrait reference when supplied, guiding the character’s appearance. Required for image-01-live, optional for image-01; only type=character is supported. Minimum items `1` Maximum items `1`
[]`body.subject_reference[]`oneOf · 2	`object`	—	Required fields `type, image_file`
Portrait subject type`body.subject_reference[].type`oneOf · 2	`string`	—	Required in the reference item and fixed to character. It does not add support for a list of arbitrary subject types. Fixed value `character`
Portrait image file`body.subject_reference[].image_file`oneOf · 2	`string`	—	Required nonempty public image URL or base64 data URL. JPG/JPEG/PNG under 10 MB; a single front-facing portrait is recommended rather than a guaranteed identity match. Minimum length `1`

View raw JSON

{
  "type": "object",
  "required": [],
  "properties": {
    "body": {
      "oneOf": [
        {
          "type": "object",
          "required": [
            "model",
            "prompt"
          ],
          "properties": {
            "n": {
              "type": "integer",
              "default": 1,
              "maximum": 9,
              "minimum": 1,
              "description": "Number of images to generate (1-9). Drives the per-result charge."
            },
            "seed": {
              "type": "integer",
              "maximum": 9007199254740991,
              "minimum": -9007199254740991,
              "description": "Random seed for reproducible images."
            },
            "model": {
              "type": "string",
              "const": "image-01",
              "description": "Image model."
            },
            "width": {
              "type": "integer",
              "maximum": 2048,
              "minimum": 512,
              "description": "Width px (image-01 only; set with height, divisible by 8)."
            },
            "height": {
              "type": "integer",
              "maximum": 2048,
              "minimum": 512,
              "description": "Height px (image-01 only; set with width, divisible by 8)."
            },
            "prompt": {
              "type": "string",
              "maxLength": 1500,
              "minLength": 1,
              "description": "Text description of the image (max 1500). Required."
            },
            "aspect_ratio": {
              "enum": [
                "1:1",
                "16:9",
                "4:3",
                "3:2",
                "2:3",
                "3:4",
                "9:16",
                "21:9"
              ],
              "type": "string",
              "default": "1:1",
              "description": "Aspect ratio. Takes priority over width/height if both given."
            },
            "response_format": {
              "enum": [
                "url",
                "base64"
              ],
              "type": "string",
              "default": "url",
              "description": "'url' links expire in 24h; 'base64' is inline (no expiry)."
            },
            "prompt_optimizer": {
              "type": "boolean",
              "default": false,
              "description": "Enable automatic prompt optimization."
            },
            "subject_reference": {
              "type": "array",
              "items": {
                "type": "object",
                "required": [
                  "type",
                  "image_file"
                ],
                "properties": {
                  "type": {
                    "type": "string",
                    "const": "character",
                    "description": "Subject type. Only `character` (portrait) is supported."
                  },
                  "image_file": {
                    "type": "string",
                    "minLength": 1,
                    "description": "Reference image: public URL or base64 data URL (data:image/jpeg;base64,...). JPG/JPEG/PNG, <10MB. A single front-facing portrait works best."
                  }
                },
                "additionalProperties": false
              },
              "maxItems": 1,
              "minItems": 1,
              "description": "Optional subject reference — supply for image-to-image."
            }
          },
          "additionalProperties": false
        },
        {
          "type": "object",
          "required": [
            "model",
            "prompt",
            "subject_reference"
          ],
          "properties": {
            "n": {
              "type": "integer",
              "default": 1,
              "maximum": 9,
              "minimum": 1,
              "description": "Number of images to generate (1-9). Drives the per-result charge."
            },
            "seed": {
              "type": "integer",
              "maximum": 9007199254740991,
              "minimum": -9007199254740991,
              "description": "Random seed for reproducible images."
            },
            "model": {
              "type": "string",
              "const": "image-01-live",
              "description": "Image-to-image (portrait) model."
            },
            "prompt": {
              "type": "string",
              "maxLength": 1500,
              "minLength": 1,
              "description": "Text description of the image (max 1500). Required."
            },
            "aspect_ratio": {
              "enum": [
                "1:1",
                "16:9",
                "4:3",
                "3:2",
                "2:3",
                "3:4",
                "9:16",
                "21:9"
              ],
              "type": "string",
              "default": "1:1",
              "description": "Aspect ratio. Takes priority over width/height if both given."
            },
            "response_format": {
              "enum": [
                "url",
                "base64"
              ],
              "type": "string",
              "default": "url",
              "description": "'url' links expire in 24h; 'base64' is inline (no expiry)."
            },
            "prompt_optimizer": {
              "type": "boolean",
              "default": false,
              "description": "Enable automatic prompt optimization."
            },
            "subject_reference": {
              "type": "array",
              "items": {
                "type": "object",
                "required": [
                  "type",
                  "image_file"
                ],
                "properties": {
                  "type": {
                    "type": "string",
                    "const": "character",
                    "description": "Subject type. Only `character` (portrait) is supported."
                  },
                  "image_file": {
                    "type": "string",
                    "minLength": 1,
                    "description": "Reference image: public URL or base64 data URL (data:image/jpeg;base64,...). JPG/JPEG/PNG, <10MB. A single front-facing portrait works best."
                  }
                },
                "additionalProperties": false
              },
              "maxItems": 1,
              "minItems": 1,
              "description": "Subject reference for image-to-image generation."
            }
          },
          "additionalProperties": false
        }
      ],
      "$schema": "https://json-schema.org/draft/2020-12/schema",
      "~standard": {
        "vendor": "zod",
        "version": 1,
        "jsonSchema": {}
      }
    }
  },
  "additionalProperties": false
}

Result schema

No fixed output contract is available. Field types reflect saved result evidence; individual responses can differ.

Result schema
Field / Path	Type	Meaning
Supplier trace ID`result.id`	`string`	Trace identifier for tracking the supplier image-generation request. Keep it with the returned image content; it is separate from the Tool run ID, prompt and random seed.
Returned image content`result.data`	`object`	Object carrying actual generated image content. Preserve the supplied result instead of inventing image dimensions, quality scores or unavailable metadata.
Inline generated images`result.data.image_base64`	`array`	Array of returned base64-encoded images; its length is the actual returned quantity. Decode each string to image bytes for display or file saving: text alone is not an image file. Retain the raw data and actual decoded image representation.

Additional notes

Create a still image from text. One portrait link is optional and is used only to guide the appearance of one person. A front-facing portrait may help, but an exact likeness is not guaranteed.

Choose a square, wide, or tall image. You can request one to nine images; asking for more can increase the cost.

A download link stays available for 24 hours. Save the image while the link works. If you need to see the result again, reuse the saved image instead of requesting a new generation.

Review the actual image before sharing it. A successful request alone does not confirm that every detail matches the description.

MiniMax image generation guide

MiniMax portrait image guide

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

No. A portrait photo is optional. Describe the image in words; add one portrait link only if you want to guide a person's appearance.

AI Portrait Generator · MiniMax

AI Portrait Generator · MiniMax

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Describe the picture

Choose the shape or share a photo

Save the finished image

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Describe the picture

Choose the shape or share a photo

Save the finished image

Questions about this tool.

Do I need to provide a portrait photo?

What kind of portrait link works?

Can I choose the image shape?

How many images can I request?

Will the image match every detail?

How long can I download my image?