Photo to Video AI · MiniMax H3

Animate a supplied photo with one motion prompt, MiniMax-H3, a whole-number duration of 4–15 seconds and 768P or 2K. Successful generation provides a video download link. Use a first frame and optional paired last frame, or choose a separate reference-media mode.

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-minimax, ID 1762e14f-72d2-4a03-9908-a83051d26b9a. Reuse the existing Toolmu connection; when needed, connect through https://toolmu.com/SKILL.md. Read this exact Tool’s current availability, inputs and customer credit ceiling, and use the same account’s established authorization, balance and budget. Obtain one 1–7,000-character scene prompt, an integer duration from 4 to 15 seconds and a resolution choice. Prefer 768P or 2K, which current MiniMax H3 documentation lists; explain that 480P is accepted input here but its delivery is unverified. Set model to MiniMax-H3. For photo animation, obtain a compatible public HTTPS image URL and assign first_frame; add last_frame only with a first frame. For text alone, choose a concrete supported ratio. For reference generation, use reference_image/video/audio roles without frame roles; audio needs a reference image or video. Validate media requirements and the combined reference count; prefer HTTPS URLs for large files. Follow Toolmu’s authorized run and spending workflow, and recover the original request if submission is uncertain. On success, retrieve and promptly save the video, retain returned settings and report the actual status and confirmed charge. Do not promise a finished video from task acceptance alone.

Task:
Turn [a link that opens the photo itself without asking you to sign in] into a short video. Show [gentle ripples moving across a lakeside scene] while the camera [slowly moves forward]. Make it [5 seconds] long and choose [clearer picture quality]. The video will keep the photo’s shape.

Paste into your connected AI's conversation.

Connection guide

Up to 4.872 credits / call

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

MiniMax

Photo to Video AI · MiniMax H3

You receive

Provide a photo as the first frame, one scene-and-motion prompt, MiniMax-H3, a 4–15-second duration and 768P or 2K to request a short animated video. After successful generation, retrieve and save the video from its download link. An optional paired last frame supplies an ending boundary; reference images, clips or audio belong to a separate generation mode.

Start with your task

Turn [a link that opens the photo itself without asking you to sign in] into a short video. Show [gentle ripples moving across a lakeside scene] while the camera [slowly moves forward]. Make it [5 seconds] long and choose [clearer picture quality]. The video will keep the photo’s shape.

What to know

Frame images and reference media cannot be combined. A last frame requires a first frame, and audio cannot be the only reference. Current H3 documentation lists 768P and 2K; the Tool also accepts 480P input, but its video delivery is unverified. Download links are time limited, so save the video promptly.

Inputs & outputs

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

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Video-generation inputs`body`	`object`	Required	Required object containing the fixed model, one scene prompt, chosen resolution and duration, plus the selected mode’s frame or reference media. Required fields `model, content, resolution, duration`
Generation model`body.model`	`string`	Required	Required model identifier: MiniMax-H3. It selects this H3 generation request. Fixed value `MiniMax-H3`
Video aspect ratio`body.ratio`	`string`	Optional	Width-to-height ratio: adaptive, 21:9, 16:9, 4:3, 1:1, 3:4 or 9:16. Text-only requires an explicit nonadaptive value. Frame-image mode follows the image and resolves to adaptive; reference mode defaults to adaptive but accepts a concrete ratio. Options `adaptive, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16`
Prompt and media items`body.content`	`array`	Required	Required list with exactly one nonempty text item. Frame mode uses one first frame and an optional paired last frame, OR choose reference media; those families cannot mix. References allow up to nine images, three videos and three audio files, with at most twelve files combined. Audio needs a reference image or video. Minimum items `1`
[]`body.content[]`	`object`	—
Scene and motion prompt`body.content[].text`oneOf · 1	`string`	—	Required on the single text item: 1–7,000 characters describing the desired scene, action and camera movement. Every supported mode needs this text, even with media. Minimum length `1` Maximum length `7000`
Item kind`body.content[].type`oneOf · 1	`string`	—	Required tag: text for the scene prompt, image_url for a frame or reference image, video_url for a reference clip, or audio_url for reference audio. Fixed value `text`
Frame or reference role`body.content[].role`oneOf · 2	`string`	—	first_frame supplies the opening image; last_frame supplies a paired ending image and requires first_frame here. One image with omitted role is treated as first_frame. Reference roles are reference_image, reference_video and reference_audio; video and audio items require their matching role. Options `first_frame, last_frame, reference_image`
Item kind`body.content[].type`oneOf · 2	`string`	—	Required tag: text for the scene prompt, image_url for a frame or reference image, video_url for a reference clip, or audio_url for reference audio. Fixed value `image_url`
Frame or reference image source`body.content[].image_url`oneOf · 2	`object`	—	Required object on an image_url item. Its url identifies the image used as the selected frame position or as a reference for generation. Required fields `url`
Image address`body.content[].image_url.url`oneOf · 2	`string`	—	Nonempty public HTTPS URL or data URI; prefer HTTPS for large files. JPG/JPEG/PNG/WEBP/HEIC/HEIF, at most 30 MB per image, 256–5,760 pixels per side and width-to-height ratio 0.4–2.5. mm_file:// is not accepted. Minimum length `1`
Frame or reference role`body.content[].role`oneOf · 3	`string`	—	first_frame supplies the opening image; last_frame supplies a paired ending image and requires first_frame here. One image with omitted role is treated as first_frame. Reference roles are reference_image, reference_video and reference_audio; video and audio items require their matching role. Fixed value `reference_video`
Item kind`body.content[].type`oneOf · 3	`string`	—	Required tag: text for the scene prompt, image_url for a frame or reference image, video_url for a reference clip, or audio_url for reference audio. Fixed value `video_url`
Reference video source`body.content[].video_url`oneOf · 3	`object`	—	Required object on a video_url item in reference mode. Its url provides a clip to guide the requested motion or camera work, with role reference_video. Required fields `url`
Reference clip address`body.content[].video_url.url`oneOf · 3	`string`	—	Nonempty public HTTPS URL or data URI; prefer HTTPS. MP4/MOV with H.264/AVC or H.265/HEVC video, and AAC/MP3 if audio is present; at most 50 MB per clip. Each clip is 2–15 seconds and all reference clips total at most 15 seconds. Dimensions: 256–5,760 pixels per side; ratio 0.4–2.5; frame rate 23.976–60 fps. mm_file:// is not accepted. Minimum length `1`
Frame or reference role`body.content[].role`oneOf · 4	`string`	—	first_frame supplies the opening image; last_frame supplies a paired ending image and requires first_frame here. One image with omitted role is treated as first_frame. Reference roles are reference_image, reference_video and reference_audio; video and audio items require their matching role. Fixed value `reference_audio`
Item kind`body.content[].type`oneOf · 4	`string`	—	Required tag: text for the scene prompt, image_url for a frame or reference image, video_url for a reference clip, or audio_url for reference audio. Fixed value `audio_url`
Reference audio source`body.content[].audio_url`oneOf · 4	`object`	—	Required object on an audio_url item, with role reference_audio. Its url supplies audio for reference generation alongside a reference image or video. Required fields `url`
Reference audio address`body.content[].audio_url.url`oneOf · 4	`string`	—	Nonempty public HTTPS URL or data URI; prefer HTTPS. WAV/MP3, at most 15 MB per file. Each clip is 2–15 seconds and all reference audio totals at most 15 seconds. Audio cannot be the only reference; mm_file:// is not accepted. Minimum length `1`
Requested clip length`body.duration`	`integer`	Required	Required whole-number output length from 4 through 15 seconds. This sets the requested clip length, separate from actual metered usage and charge. Minimum `4` Maximum `15`
Generated-video resolution`body.resolution`	`string`	Required	The input accepts 480P, 768P or 2K, with no preset. Current MiniMax H3 documentation lists 768P and 2K; choose either for the documented output choices. 480P video delivery remains unverified despite accepted input. Options `480P, 768P, 2K`

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",
          "description": "Fixed to MiniMax-H3. 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": 4,
          "description": "Output length in seconds, 4-15."
        },
        "resolution": {
          "enum": [
            "480P",
            "768P",
            "2K"
          ],
          "type": "string",
          "description": "Output resolution. Drives the per-second rate: 480P $0.038/s, 768P $0.08/s, 2K $0.13/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 photo supplies the starting image; the text prompt describes the action and camera movement to generate. A paired last frame adds an ending boundary. Every request still needs one text prompt. With no image, use text-to-video and specify a concrete aspect ratio; with frame images, the image determines the ratio.

Reference generation uses a separate family of inputs: images can guide appearance, clips can guide motion or camera work, and audio can provide an audio reference. These roles cannot be mixed with first/last frames. Audio needs at least a reference image or video. Keep within nine reference images, three clips and three audio files, with no more than twelve reference files combined; matching media requirements still apply.

MiniMax describes generation as asynchronous: an accepted task is not yet a finished video. After success, the video-task result supplies a time-limited download link. Save the video promptly. A supplier task identifier tracks that generation and is separate from the Toolmu run identifier used by your Agent.

Where returned, the task’s model, resolution, duration and aspect ratio describe the generated video. MiniMax defines task creation and update timestamps as Unix seconds. Output seconds measure the produced clip; input seconds measure reference-video input, and total seconds combine metered input and output. The image count records supplied images. Keep zero and missing values distinct, and treat these usage measures separately from Toolmu’s confirmed credit charge.

MiniMax H3 generation modes and media requirements

MiniMax video task and download-link definitions

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

No. This Tool requires last_frame to be paired with first_frame. Keep one text prompt even when both frame images are supplied.

Photo to Video AI · MiniMax H3

Photo to Video AI · MiniMax H3

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose a photo and describe the movement

Choose video length and 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 photo and describe the movement

Choose video length and clarity

Open and save your video

Questions about this tool.

Can I supply only an ending image?

Can I mix frame images with reference files?

Can audio be my only reference?

Which H3 resolution should I choose?