Image to Video · MiniMax H3 Max

Bring a first-frame image into motion with one text description, MiniMax-H3-Max, a choice of 480P or 768P and an integer duration of 5–15 seconds. Add a paired last frame when needed, or use the separate text-only or reference-media modes.

Your AI

Use with your AI

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

CodexClaude CodeCursorWindsurfCline

View instructions

Use Toolmu Tool minimax-video-h3-max-minimax, ID 416034dc-cd2c-4ea9-99ea-79f6f2bf4b13. Keep the existing Toolmu connection; if none exists, follow https://toolmu.com/SKILL.md. Read the exact current Tool contract, availability and pricing, then check identity, balance and established budget in the same environment. Ask only for missing scene text, duration, resolution choice and any selected image/reference source. Use model MiniMax-H3-Max, explicitly choose 480P or 768P and an integer duration of 5–15 seconds. Build exactly one nonempty text item. For image-to-video, use a first frame and add a last frame only with that first frame; the image determines the ratio. Text-only requires an explicit non-adaptive ratio. Reference-media mode uses its own image/video/audio roles and cannot be mixed with first/last frames; audio cannot be the sole reference. Check all selected media limits before submission. Use one persisted idempotency key and an authorized maxCredits bound. Follow the same run and recover an uncertain submission rather than paying for another generation. Present only the actual returned video information and confirmed credits, preserving absent fields and user-supplied media identities. Do not claim that 768P guarantees better results or that every request completes at the same speed.

Task:
Turn [a link to a picture that opens directly without signing in] into a short video. Show [a paper boat floating down a stream] while the camera [slowly follows it]. Make it [5 seconds] long and choose [standard or clearer picture quality].

Paste into your connected AI's conversation.

Connection guide

Up to 4.014 credits / call

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

MiniMax

Image to Video · MiniMax H3 Max

You receive

Provide a first-frame image, one nonempty video description, MiniMax-H3-Max, 480P or 768P and a 5–15-second duration to create a short video. You can instead use text alone with an explicit non-adaptive ratio, or compatible reference media.

Start with your task

Turn [a link to a picture that opens directly without signing in] into a short video. Show [a paper boat floating down a stream] while the camera [slowly follows it]. Make it [5 seconds] long and choose [standard or clearer picture quality].

What to know

A last frame requires a first frame. Frame images and reference-media inputs cannot be mixed; audio needs a visual reference. Mixed reference inputs are capped at 12 files. Resolution choice does not guarantee reference fidelity or audiovisual quality.

Inputs & outputs

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

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Generation request`body`	`object`	Required	Mandatory object for the model, content array, resolution and duration, with ratio and media sources supplied as the chosen mode requires. Required fields `model, content, resolution, duration`
Model selection`body.model`	`string`	Required	Selects the video-generation model. The required value is fixed to MiniMax-H3-Max. Fixed value `MiniMax-H3-Max`
Width-to-height ratio`body.ratio`	`string`	Optional	Supported values are adaptive, 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16. A text-only request must explicitly use a non-adaptive value. Image-to-video resolves to adaptive from the image; references default to adaptive but can use a concrete ratio. Options `adaptive, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16`
Scene and reference content`body.content`	`array`	Required	Required array with exactly one nonempty text item. Add first/last-frame images or reference-media items, keeping these two families separate. Minimum items `1`
[]`body.content[]`	`object`	—
Motion description`body.content[].text`oneOf · 1	`string`	—	The text item’s required video prompt, 1–7,000 characters. It describes the intended scene or movement and remains required with any frame or reference media. Minimum length `1` Maximum length `7000`
Item type`body.content[].type`oneOf · 1	`string`	—	Identifies each content item as text, image_url, video_url or audio_url. The chosen tag determines which text or URL object must be supplied. Fixed value `text`
Frame or reference role`body.content[].role`oneOf · 2	`string`	—	Image roles are first_frame, last_frame and reference_image; a single image with no role defaults to first_frame, while last_frame requires a first frame. Video and audio items require reference_video and reference_audio. Reference limits are nine images, three videos and three audios, with at most 12 files combined; audio must have a visual reference. Options `first_frame, last_frame, reference_image`
Item type`body.content[].type`oneOf · 2	`string`	—	Identifies each content item as text, image_url, video_url or audio_url. The chosen tag determines which text or URL object must be supplied. Fixed value `image_url`
Image location object`body.content[].image_url`oneOf · 2	`object`	—	Required for an image_url item. Its url identifies the frame or visual-reference image selected by the item’s role. Required fields `url`
Frame or image URL`body.content[].image_url.url`oneOf · 2	`string`	—	Provide a nonempty public HTTPS URL or data URI; prefer HTTPS for large images. Supported JPG/JPEG/PNG/WEBP/HEIC/HEIF images are at most 30 MB each, 256–5,760 pixels per side and 0.4–2.5 in width-to-height ratio. mm_file:// references are not supported. Minimum length `1`
Frame or reference role`body.content[].role`oneOf · 3	`string`	—	Image roles are first_frame, last_frame and reference_image; a single image with no role defaults to first_frame, while last_frame requires a first frame. Video and audio items require reference_video and reference_audio. Reference limits are nine images, three videos and three audios, with at most 12 files combined; audio must have a visual reference. Fixed value `reference_video`
Item type`body.content[].type`oneOf · 3	`string`	—	Identifies each content item as text, image_url, video_url or audio_url. The chosen tag determines which text or URL object must be supplied. Fixed value `video_url`
Video reference location`body.content[].video_url`oneOf · 3	`object`	—	The required media-location object for a video_url item. Pair it with reference_video in reference-media mode. Required fields `url`
Video reference URL`body.content[].video_url.url`oneOf · 3	`string`	—	Use a nonempty public HTTPS URL or data URI, preferring HTTPS for large clips. Reference video supports MP4/MOV with H.264/H.265 and AAC/MP3 in-video audio. Each clip is at most 50 MB and 2–15 seconds, with at most 15 seconds total. Width and height are 256–5,760 pixels, and aspect ratio is 0.4–2.5. mm_file:// is unsupported. Minimum length `1`
Frame or reference role`body.content[].role`oneOf · 4	`string`	—	Image roles are first_frame, last_frame and reference_image; a single image with no role defaults to first_frame, while last_frame requires a first frame. Video and audio items require reference_video and reference_audio. Reference limits are nine images, three videos and three audios, with at most 12 files combined; audio must have a visual reference. Fixed value `reference_audio`
Item type`body.content[].type`oneOf · 4	`string`	—	Identifies each content item as text, image_url, video_url or audio_url. The chosen tag determines which text or URL object must be supplied. Fixed value `audio_url`
Audio reference location`body.content[].audio_url`oneOf · 4	`object`	—	The audio_url item’s required location object. Use role reference_audio together with at least an image or video reference; audio alone is not a supported reference request. Required fields `url`
Audio reference URL`body.content[].audio_url.url`oneOf · 4	`string`	—	Use a nonempty public HTTPS URL or data URI for WAV/MP3, preferring HTTPS for large audio files. Each clip is at most 15 MB and 2–15 seconds, with at most 15 seconds total audio reference. mm_file:// is not supported. Minimum length `1`
Video duration`body.duration`	`integer`	Required	Required whole-number duration, from 5 to 15 seconds inclusive. Select the requested length separately from resolution. Minimum `5` Maximum `15`
Video resolution`body.resolution`	`string`	Required	Choose the generated video’s resolution explicitly: 480P or 768P. This required field has no preset value. Options `480P, 768P`

View raw JSON

{
  "type": "object",
  "required": [
    "body"
  ],
  "properties": {
    "body": {
      "type": "object",
      "$schema": "https://json-schema.org/draft/2020-12/schema",
      "required": [
        "model",
        "content",
        "resolution",
        "duration"
      ],
      "~standard": {
        "vendor": "zod",
        "version": 1,
        "jsonSchema": {}
      },
      "properties": {
        "model": {
          "type": "string",
          "const": "MiniMax-H3-Max",
          "description": "Fixed to MiniMax-H3-Max. Required."
        },
        "ratio": {
          "enum": [
            "adaptive",
            "21:9",
            "16:9",
            "4:3",
            "1:1",
            "3:4",
            "9:16"
          ],
          "type": "string",
          "description": "Aspect ratio. REQUIRED and must not be adaptive for text-to-video. Image-to-video always resolves to adaptive (the input image decides). Reference-to-video defaults to adaptive but accepts a concrete ratio."
        },
        "content": {
          "type": "array",
          "items": {
            "oneOf": [
              {
                "type": "object",
                "required": [
                  "type",
                  "text"
                ],
                "properties": {
                  "text": {
                    "type": "string",
                    "maxLength": 7000,
                    "minLength": 1,
                    "description": "Prompt describing the video (max 7000 characters). REQUIRED in every request, in every scenario."
                  },
                  "type": {
                    "type": "string",
                    "const": "text"
                  }
                },
                "additionalProperties": false
              },
              {
                "type": "object",
                "required": [
                  "type",
                  "image_url"
                ],
                "properties": {
                  "role": {
                    "enum": [
                      "first_frame",
                      "last_frame",
                      "reference_image"
                    ],
                    "type": "string",
                    "description": "first_frame (default when a single image carries no role), last_frame (must be paired with first_frame), or reference_image (reference-to-video, max 9). Formats: JPG, JPEG, PNG, WEBP, HEIC, HEIF; 256-5760px per side; aspect ratio 0.4-2.5."
                  },
                  "type": {
                    "type": "string",
                    "const": "image_url"
                  },
                  "image_url": {
                    "type": "object",
                    "required": [
                      "url"
                    ],
                    "properties": {
                      "url": {
                        "type": "string",
                        "minLength": 1,
                        "description": "Public HTTPS URL, or a data: URI (data:image/jpeg;base64,...). MiniMax mm_file:// references are not supported. Limits: image <=30MB, reference video <=50MB, reference audio <=15MB, whole request body <=64MB — prefer URLs over data: URIs for video."
                      }
                    },
                    "additionalProperties": false
                  }
                },
                "additionalProperties": false
              },
              {
                "type": "object",
                "required": [
                  "type",
                  "video_url",
                  "role"
                ],
                "properties": {
                  "role": {
                    "type": "string",
                    "const": "reference_video",
                    "description": "Reference video (reference-to-video only, max 3). MP4/MOV, H.264 or H.265, 2-15s per clip and <=15s total."
                  },
                  "type": {
                    "type": "string",
                    "const": "video_url"
                  },
                  "video_url": {
                    "type": "object",
                    "required": [
                      "url"
                    ],
                    "properties": {
                      "url": {
                        "type": "string",
                        "minLength": 1,
                        "description": "Public HTTPS URL, or a data: URI (data:image/jpeg;base64,...). MiniMax mm_file:// references are not supported. Limits: image <=30MB, reference video <=50MB, reference audio <=15MB, whole request body <=64MB — prefer URLs over data: URIs for video."
                      }
                    },
                    "additionalProperties": false
                  }
                },
                "additionalProperties": false
              },
              {
                "type": "object",
                "required": [
                  "type",
                  "audio_url",
                  "role"
                ],
                "properties": {
                  "role": {
                    "type": "string",
                    "const": "reference_audio",
                    "description": "Reference audio (reference-to-video only, max 3, cannot be the only reference). WAV/MP3, 2-15s per clip and <=15s total."
                  },
                  "type": {
                    "type": "string",
                    "const": "audio_url"
                  },
                  "audio_url": {
                    "type": "object",
                    "required": [
                      "url"
                    ],
                    "properties": {
                      "url": {
                        "type": "string",
                        "minLength": 1,
                        "description": "Public HTTPS URL, or a data: URI (data:image/jpeg;base64,...). MiniMax mm_file:// references are not supported. Limits: image <=30MB, reference video <=50MB, reference audio <=15MB, whole request body <=64MB — prefer URLs over data: URIs for video."
                      }
                    },
                    "additionalProperties": false
                  }
                },
                "additionalProperties": false
              }
            ]
          },
          "minItems": 1,
          "description": "Multimodal input. Every request needs exactly one non-empty text item. Add first_frame/last_frame images for image-to-video, OR reference_image/reference_video/reference_audio items for reference-to-video — the two families cannot be mixed."
        },
        "duration": {
          "type": "integer",
          "maximum": 15,
          "minimum": 5,
          "description": "Output length in seconds, 5-15."
        },
        "resolution": {
          "enum": [
            "480P",
            "768P"
          ],
          "type": "string",
          "description": "Output resolution. Drives the per-second rate: 480P $0.05/s, 768P $0.08/s. The USD amounts in this input description are upstream supplier values; Toolmu charges the current catalog credits."
        }
      },
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}

Result schema

Output fields have not been documented yet.

Additional notes

A first frame supplies the opening image of an image-to-video request, while a paired last frame supplies its ending boundary. Describe the desired motion in one text item, then select the output resolution and duration.

Text-only and reference-media modes share the fixed model, resolution choices and duration range. Text-only needs a concrete aspect ratio. References use their own image/video/audio roles; audio requires a visual reference. The individual maxima of nine images, three videos and three audios are also subject to a combined limit of 12 reference files.

For a completed video task, MiniMax defines content.url as a time-limited download address. Save the actual video promptly when a link is returned. Supplier task identity is separate from the Toolmu run ID; created_at and updated_at are Unix seconds for task creation and latest status update.

Supplier status describes generation progress, with succeeded indicating completion. Returned usage can distinguish reference-video input seconds, generated output seconds and their metered total, plus image counts. Keep any returned settings and usage attached to the result, without treating those counts as the confirmed Toolmu credit charge.

MiniMax H3 Max generation reference

MiniMax video output and task definitions

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

No. Explicitly choose 480P or 768P along with the required model, duration and content. There is no preset resolution. Select it before the Agent checks the current price and submits your generation.

Image to Video · MiniMax H3 Max

Image to Video · MiniMax H3 Max

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose a picture and describe the motion

Choose the length and picture clarity

Open and save your video

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose a picture and describe the motion

Choose the length and picture clarity

Open and save your video

Questions about this tool.

Can I leave the resolution unspecified?

Can I supply only the ending image?

Do reference files replace the text prompt?

Is every duration and resolution charged identically?