MiniMax H3 Video (workflow_id)

MiniMax H3 uses the OpenAI-compatible asynchronous video API: submit a task → poll its status → download the video. Each request selects a workflow with workflow_id; the workflow decides which reference media are allowed and which sizes are supported.

POST https://api.4allapi.com/v1/videos
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json   (multipart/form-data when uploading local files)

Use the model name minimax-h3 (as shown in the console model marketplace). MiniMax H3 supports the OpenAI protocol only, not Gemini predictLongRunning.

Choosing a workflow

`workflow_id`	Reference media	Size
`text-to-video`	No reference media; text only	Up to 1080P
`fl2v`	1–2 images; with two, they are the first and last frame in order; no video or audio	Up to 1080P
`multi-reference`	Optional; up to 9 images, up to 3 videos and 3 audios	Up to 1080P
`cf-fl2v`	Same as `fl2v`	Must be 2K or 4K
`cf-multi-reference`	Same as `multi-reference`	Must be 2K or 4K

Quick choice: text only → text-to-video; 1–2 images as first (and last) frame → fl2v; several images, reference videos or music → multi-reference. For 2K/4K output use the matching cf- workflow.

Requests that break the workflow rules (e.g. images with text-to-video, no image for fl2v, too many items, a cf- workflow below 2K) are rejected with 400 before submission and are not charged.

Request parameters

Parameter	Type	Required	Description
`model`	string	yes	`minimax-h3`
`prompt`	string	yes	Video description
`workflow_id`	string	yes	See the table above
`seconds`	integer	no	4–15 seconds, default 10
`resolution`	string	no	`480p`, `768p`, `1080p`; `2k`, `4k` only with `cf-` workflows
`aspect_ratio`	string	no	`16:9` (default), `9:16`, `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `21:9`
`images`	string[]	per workflow	Public image URLs; in `fl2v` the 1st is the first frame and the 2nd the last frame
`reference_videos`	string[]	no	Public video URLs, up to 3 (`multi-reference` only)
`reference_audios`	string[]	no	Public audio URLs, up to 3 (`multi-reference` only)
`prompt_enhance`	boolean	no	Prompt enhancement, on by default

resolution and aspect_ratio are mapped to the upstream’s standard canvas sizes, e.g. 768P 16:9 = 1376x768, 1080P 16:9 = 1920x1088, 1080P 9:16 = 1088x1920. 2K/4K are submitted as 2K/4K plus the aspect ratio.

Media limits: images JPG/PNG/WEBP/HEIC/HEIF, ≤ 30 MB each; videos MP4/MOV, ≤ 50 MB and 2–15 s each; audio WAV/MP3, ≤ 15 MB and 2–15 s each; total video and total audio length ≤ 15 s each. JSON requests must use public http(s) URLs (no Base64 / data URLs); upload local files with multipart.

Example: text-to-video

curl https://api.4allapi.com/v1/videos \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3",
    "prompt": "一位身穿红色连衣裙的舞者在明亮舞台上自然起舞，镜头平稳推进，电影级光影",
    "workflow_id": "text-to-video",
    "seconds": 10,
    "resolution": "1080p",
    "aspect_ratio": "16:9"
  }'

Example: first/last frame (fl2v)

curl https://api.4allapi.com/v1/videos \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3",
    "prompt": "从首帧平滑过渡到尾帧，镜头运动自然稳定",
    "workflow_id": "fl2v",
    "seconds": 10,
    "resolution": "768p",
    "aspect_ratio": "9:16",
    "images": [
      "https://cdn.example.com/first-frame.png",
      "https://cdn.example.com/last-frame.png"
    ]
  }'

With a single image, it is used as the first frame.

Example: multi-reference

curl https://api.4allapi.com/v1/videos \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3",
    "prompt": "将参考人物、动作和音乐融合成一段自然舞蹈视频",
    "workflow_id": "multi-reference",
    "seconds": 8,
    "resolution": "1080p",
    "aspect_ratio": "16:9",
    "images": ["https://cdn.example.com/person.png", "https://cdn.example.com/outfit.png"],
    "reference_videos": ["https://cdn.example.com/dance.mp4"],
    "reference_audios": ["https://cdn.example.com/music.mp3"]
  }'

Example: 4K first/last frame (cf-fl2v)

curl https://api.4allapi.com/v1/videos \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3",
    "prompt": "从首帧平滑过渡到尾帧，镜头运动自然稳定",
    "workflow_id": "cf-fl2v",
    "seconds": 10,
    "resolution": "4k",
    "aspect_ratio": "16:9",
    "images": ["https://cdn.example.com/first-frame.png", "https://cdn.example.com/last-frame.png"]
  }'

Uploading local files (multipart)

Repeat image[] for images, video[] for videos and audio[] for audio; other parameters are plain form fields. Files are stored on our media host before being sent upstream.

curl https://api.4allapi.com/v1/videos \
  -H "Authorization: Bearer $API_KEY" \
  -F "model=minimax-h3" \
  -F "prompt=从首帧平滑过渡到尾帧" \
  -F "workflow_id=fl2v" \
  -F "seconds=10" \
  -F "resolution=768p" \
  -F "aspect_ratio=16:9" \
  -F "image[]=@first.png;type=image/png" \
  -F "image[]=@last.png;type=image/png"

Full Python example

import time
import requests

BASE_URL = "https://api.4allapi.com"
HEADERS = {"Authorization": "Bearer YOUR_API_KEY"}

body = {
    "model": "minimax-h3",
    "prompt": "从首帧平滑过渡到尾帧，镜头运动自然稳定",
    "workflow_id": "fl2v",
    "seconds": 10,
    "resolution": "1080p",
    "aspect_ratio": "16:9",
    "images": [
        "https://cdn.example.com/first-frame.png",
        "https://cdn.example.com/last-frame.png",
    ],
}

task = requests.post(f"{BASE_URL}/v1/videos", headers=HEADERS, json=body).json()
task_id = task["id"]

while task.get("status") not in ("completed", "failed"):
    time.sleep(5)
    task = requests.get(f"{BASE_URL}/v1/videos/{task_id}", headers=HEADERS).json()
    print(task.get("status"), task.get("progress"))

if task["status"] == "completed":
    video = requests.get(f"{BASE_URL}/v1/videos/{task_id}/content", headers=HEADERS)
    video.raise_for_status()
    open("output.mp4", "wb").write(video.content)

Query and download

# Status: queued / in_progress / completed / failed
curl https://api.4allapi.com/v1/videos/<TASK_ID> -H "Authorization: Bearer $API_KEY"

# Download the video (same API key)
curl -L https://api.4allapi.com/v1/videos/<TASK_ID>/content \
  -H "Authorization: Bearer $API_KEY" -o output.mp4

Common errors

Message	Cause
`workflow text-to-video is text-to-video only and accepts no reference media`	Reference media sent with `text-to-video`
`workflow fl2v needs 1-2 images; got 0`	`fl2v` / `cf-fl2v` without images or with more than 2
`workflow cf-fl2v requires size 2K or 4K`	`cf-` workflow with a `resolution` other than `2k`/`4k`
`2K/4K requires a cf- workflow`	`2k`/`4k` with a non-`cf-` workflow
`seconds must be between 4 and 15`	Duration outside 4–15 seconds
`unsupported video aspect ratio`	Aspect ratio not in the supported list
`workflow xxx is not configured for minimax-h3`	That workflow is not enabled for your group