MiniMax H3
Sign in to try MiniMax H3
Sign in with your Google account and top up credits to get started.
Sign in with GoogleUse via API
Base URL: https://adverraai.com/hub/api/v1 · Authenticate with Authorization: Bearer YOUR_API_KEY — create a key in your dashboard
# 1) Submit a task
curl https://adverraai.com/hub/api/v1/videos/generations \
-H "Authorization: Bearer $ADVERRA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "MiniMax-H3",
"prompt": "a cinematic shot of a city at night",
"duration": 15,
"resolution": "2K",
"aspect_ratio": "9:16"
}'
# => {"id":"task_xxx","status":"queued",...}
# 2) Check status (every 3-5 seconds until status = completed)
curl https://adverraai.com/hub/api/v1/tasks/task_xxx -H "Authorization: Bearer $ADVERRA_API_KEY"
# => {"status":"completed","output":[{"type":"video","url":"https://..."}]}
import requests, time
API = "https://adverraai.com/hub/api/v1"
H = {"Authorization": "Bearer YOUR_API_KEY"}
task = requests.post(f"{API}/videos/generations", headers=H, json={
"model": "MiniMax-H3",
"prompt": "a cinematic shot of a city at night",
"duration": 15,
"resolution": "2K",
"aspect_ratio": "9:16"
}).json()
while task["status"] not in ("completed", "failed"):
time.sleep(4)
task = requests.get(f"{API}/tasks/{task['id']}", headers=H).json()
print(task["status"], [o["url"] for o in task["output"]], task["credits_charged"])
const API = "https://adverraai.com/hub/api/v1", H = { "Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json" };
let task = await (await fetch(`${API}/videos/generations`, { method: "POST", headers: H, body: JSON.stringify({
model: "MiniMax-H3",
"prompt": "a cinematic shot of a city at night",
"duration": 15,
"resolution": "2K",
"aspect_ratio": "9:16"
}) })).json();
while (!["completed", "failed"].includes(task.status)) {
await new Promise(r => setTimeout(r, 4000));
task = await (await fetch(`${API}/tasks/${task.id}`, { headers: H })).json();
}
console.log(task.output);
Common parameters: prompt, image_urls, size / aspect_ratio, resolution, duration, n, negative_prompt, seed. Credits are reserved at an estimated price when you submit and adjusted to the actual cost when the task finishes (failed tasks are fully refunded).
Parameters for this model
Send as JSON in the request body. Fields marked * are required.
| Parameter | Type | Allowed values | Default | Description |
|---|---|---|---|---|
model * |
string | – | – | ID of the model to call — MiniMax-H3 |
prompt * |
string | – | – | Text describing what to generate |
duration |
number | 4 – 15 | 15 | Length of the result in seconds |
resolution |
string | 2K, 768P | 2K | Output resolution |
aspect_ratio |
string | 21:9, 16:9, 4:3, 1:1, 3:4, 9:16 | 9:16 | Width:height ratio of the result |
watermark |
boolean | true, false | false | Whether to add a watermark |
first_frame_image |
string (URL) | – | – | Image URL used as the first frame |
last_frame_image |
string (URL) | – | – | Image URL used as the last frame |
image_urls |
array (URL) | – | – | Reference image URLs |
video_urls |
array (URL) | – | – | Reference video URLs |
audio_urls |
array (URL) | – | – | Reference audio URLs |
About this model
MiniMax H3 is a Video model by MiniMax
MiniMax H3 is MiniMax's video model for text-to-video, image-to-video with first and last frames, and multimodal reference-to-video using images, videos and audio. It outputs 4–15 second clips at 2K or 768P with an audio track.
Preparing details in English…