# MiniMax H3

> Run MiniMax H3 by minimax online or via API. $0.08/sec (768P), $0.13/sec (2K), first 5 images free, +$0.04/extra image, no setup.

**URL:** https://inference.sh/apps/minimax/h3
**Last updated:** 2026-09-06

---

MiniMax H3 — multimodal video generation with native audio. Text-to-video, image-to-video with end_image, and reference-based generation. 2K resolution, 5-15s duration, 24fps. Prompt with timelines, audio cues, and negative lists.

- app: `minimax/h3`
- maker: [minimax](https://inference.sh/apps/minimax.md)
- category: [video](https://inference.sh/apps/category/video.md)
- input → output: text → video, image → video, reference → video
- tasks: [text to video](https://inference.sh/apps/tag/text-to-video.md), [image to video](https://inference.sh/apps/tag/image-to-video.md), [reference to video](https://inference.sh/apps/tag/reference-to-video.md), [native audio](https://inference.sh/apps/tag/native-audio.md)
- price: $0.08/sec (768P), $0.13/sec (2K), first 5 images free, +$0.04/extra image
- runs via: api, javascript and python sdk, belt cli, mcp

## h3 API Guide

### About
minimax h3 — multimodal video generation with native audio. text-to-video, image-to-video with end_image, and reference-based generation. 2k resolution, 5-15s duration, 24fps. prompt with timelines, audio cues, and negative lists.

### Get an API Key
Create one in [settings → api keys](https://app.inference.sh/settings/keys). The examples read it from `INFERENCE_API_KEY`.

```bash
export INFERENCE_API_KEY="inf_your_key"
```

### 1. Calling the API

#### Install the Client

```bash
pip install inferencesh
```

#### Run and Get Result
Submit a request and wait for the final result.

```python
import os
from inferencesh import inference

client = inference(api_key=os.environ["INFERENCE_API_KEY"])

## version @2k8vr6td pinned
result = client.run({
        "app": "minimax/h3@2k8vr6td",
        "input": {
            "prompt": "[0s-3s] Static wide shot. A lone marble camera on a pedestal in darkness, single moonbeam. [3s-6s] The lens flares to life, film strips unfurl from the reels. [6s-8s] Film strips fill the frame. Camera: locked off, no cuts. Audio: low hum building to orchestral swell at 6s. Do not add subtitles."
        }
    })

print(result["output"])
```

#### Stream Live Updates
Get real-time progress updates as the task runs.

```python
import os
from inferencesh import inference

client = inference(api_key=os.environ["INFERENCE_API_KEY"])

## version @2k8vr6td pinned
## stream=True yields updates as they arrive
for update in client.run({
        "app": "minimax/h3@2k8vr6td",
        "input": {
            "prompt": "[0s-3s] Static wide shot. A lone marble camera on a pedestal in darkness, single moonbeam. [3s-6s] The lens flares to life, film strips unfurl from the reels. [6s-8s] Film strips fill the frame. Camera: locked off, no cuts. Audio: low hum building to orchestral swell at 6s. Do not add subtitles."
        }
    }, stream=True):
    if update.get("progress"):
        print(f"progress: {update['progress']}%")
    if update.get("output"):
        print(f"output: {update['output']}")
```

#### Run in the Background
Start the task and get its id back right away, then check on it with `GET /tasks/{id}`. It is done when `status` is 10 (completed), 11 (failed) or 12 (cancelled).

```python
import os
from inferencesh import inference, TaskStatus

client = inference(api_key=os.environ["INFERENCE_API_KEY"])

## returns as soon as the task is created
task = client.run({
        "app": "minimax/h3@2k8vr6td",
        "input": {
            "prompt": "[0s-3s] Static wide shot. A lone marble camera on a pedestal in darkness, single moonbeam. [3s-6s] The lens flares to life, film strips unfurl from the reels. [6s-8s] Film strips fill the frame. Camera: locked off, no cuts. Audio: low hum building to orchestral swell at 6s. Do not add subtitles."
        }
    }, wait=False)

## later: read its current state
state = client.tasks.get(task["id"]).data
if state["status"] == TaskStatus.COMPLETED:
    print(state["output"])
```

### 2. Files

#### Automatic Upload

```python
## local file paths are automatically uploaded
result = client.run({
    "app": "minimax/h3@2k8vr6td",
    "input": {
        "image": "/path/to/local/image.png",  # detected & uploaded
        "audio": "https://example.com/audio.mp3",  # url passed through
    }
})
```

#### Manual Upload

```python
## upload and get a hosted URL
file = client.files.upload("/path/to/file.png")
print(file.uri)  # https://cloud.inference.sh/...
```

### 3. Schema

#### Input
- **prompt** (string) *required*: Video prompt. Use six-block structure: style contract, timeline with [0s-2s] markers, camera, audio cues, spelled-out text, negative list. Max 7000 chars.
- **image** (string): First frame image for image-to-video. Formats: JPG, PNG, WEBP, HEIC. Max 30MB.
- **last_image** (string): Last frame image. Requires image to be set as first frame.
- **reference_images** (array): Reference images for style/content guidance. Cannot combine with first/last frame.
- **reference_video** (string): Reference video for motion/style guidance. MP4/MOV, max 50MB, 2-15s.
- **reference_audio** (string): Reference audio for synchronized generation. WAV/MP3, max 15MB, 2-15s.
- **resolution** (string): Output resolution. 2K is the only resolution currently accepted.
- **duration** (integer): Video duration in seconds. 5-10 for text/image-to-video, 5-15 for reference-to-video.
- **ratio** (string): Aspect ratio. Text-to-video requires explicit value (not adaptive). Image/reference modes default to adaptive.

#### Output
- **video** (string): The generated video file.

## other video apps

- [VACE](https://inference.sh/apps/pruna/vace.md): $0.40/video
- [P-Video-Animate](https://inference.sh/apps/pruna/p-video-animate.md): $0.03/sec 720p, $0.06/sec 1080p
- [P-Video-Upscale](https://inference.sh/apps/pruna/p-video-upscale.md): Std: $0.01/sec 720p, $0.02 1080p, $0.04 2k, $0.08 4k; high fps 2x
- [P-Video-2-Pro-Avatar](https://inference.sh/apps/pruna/p-video-2-pro-avatar.md): $0.02/sec 480p, $0.04/sec 768p, $0.06/sec 1080p, $0.08/sec 2k, $0.12/sec 4k
- [P-Video-2-Pro-Reference](https://inference.sh/apps/pruna/p-video-2-pro-reference.md): Output: cost $0.015-$0.105/sec, speed $0.02-$0.115/sec. Ref video: $0.03-$0.06/sec. Ref img: $0.005/img. Ref audio: $0.0015-$0.0025/sec

## faq

### how much does MiniMax H3 cost?

$0.08/sec (768P), $0.13/sec (2K), first 5 images free, +$0.04/extra image. you pay per run from your inference shell credits. no subscription is required.

### how do I run MiniMax H3 via api?

call minimax/h3 with the javascript or python sdk, the rest api, or `belt app run minimax/h3` from a terminal. the api reference on this page lists every input.

### who makes MiniMax H3?

MiniMax H3 is made by minimax and runs on inference shell. every minimax app is listed at https://inference.sh/apps/minimax.

### can my agent use MiniMax H3?

yes. agents call it as a tool through the inference shell mcp server (https://api.inference.sh/mcp), the belt cli or the sdk, and pay per run like you do.

### what are the alternatives to MiniMax H3?

other video apps on inference shell include VACE, P-Video-Animate, P-Video-Upscale, P-Video-2-Pro-Avatar. they run through the same api, so switching is a one-line change.