# MiniMax H3 Max

> Run MiniMax H3 Max by falai online or via API. From $0.08 per second, no setup.

**URL:** https://inference.sh/apps/falai/minimax-h3-max
**Last updated:** 2026-09-01

---

MiniMax H3 Max — fal's post-trained H3 for stronger prompt adherence and aesthetics. Text-to-video, image-to-video with end frame, and reference-based generation. 480P/768P, 5-15s, ~3s per 5s clip.

- app: `falai/minimax-h3-max`
- maker: [falai](https://inference.sh/apps/falai.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)
- price: $0.08/sec (768P), $0.05/sec (480P)
- runs via: api, javascript and python sdk, belt cli, mcp

## minimax-h3-max API Guide

### About
minimax h3 max — fal's post-trained h3 for stronger prompt adherence and aesthetics. text-to-video, image-to-video with end frame, and reference-based generation. 480p/768p, 5-15s, ~3s per 5s clip.

### 1. Calling the API

#### Install the Client

```bash
pip install inferencesh
```

#### Setup Your API Key
Set `INFERENCE_API_KEY` as an environment variable. Get your key from settings → api keys.

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

#### Run and Get Result

```python
import os
from inferencesh import inference

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


result = client.run({
        "app": "falai/minimax-h3-max@4w1tph43",
        "input": {
            "prompt": "A cinematic tracking shot through a misty forest at dawn, golden light filtering through ancient trees."
        }
    })

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

#### Stream Live Updates

```python
import os
from inferencesh import inference

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


## stream=True yields updates as they arrive
for update in client.run({
        "app": "falai/minimax-h3-max@4w1tph43",
        "input": {
            "prompt": "A cinematic tracking shot through a misty forest at dawn, golden light filtering through ancient trees."
        }
    }, stream=True):
    if update.get("progress"):
        print(f"progress: {update['progress']}%")
    if update.get("output"):
        print(f"output: {update['output']}")
```

### 2. Authentication
The API uses API keys for authentication. See the authentication docs for detailed setup instructions.

### 3. Files

#### Automatic Upload

```python
## local file paths are automatically uploaded
result = client.run({
    "app": "falai/minimax-h3-max@4w1tph43",
    "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/...
```

### 4. Schema

#### Input
- **prompt** (string) *required*: Video prompt. Max 50000 chars. Use timeline markers like [0s-3s] for scene pacing.
- **image** (string): First frame image for image-to-video. Formats: JPG, PNG, WEBP, GIF, AVIF.
- **end_image** (string): Last frame image. Requires image to be set as first frame.
- **reference_images** (array): Reference images for style/subject guidance (max 9). Cannot combine with first/last frame.
- **reference_videos** (array): Reference videos for motion/style (max 3). MP4, 2-15s each.
- **reference_audios** (array): Reference audio clips (max 3). Requires reference image or video. 2-15s each.
- **duration** (integer): Video duration in seconds (5-15).
- **resolution** (string): Output resolution. 768P for best quality, 480P for faster/cheaper.
- **aspect_ratio** (string): Aspect ratio. Text-to-video requires explicit value (not adaptive). Reference mode defaults to adaptive.
- **prompt_expansion_mode** (string): balanced (~1s) or quality (~30s extra) for richer prompt rewriting.
- **seed** (integer): Random seed for reproducible generation.

#### Output
- **expanded_prompt** (string): The prompt after expansion, as sent to the model.
- **seed** (integer): The seed used for generation.
- **video** (string): The generated video file.

## other video apps

- [Grok Imagine Video 1.5](https://inference.sh/apps/xai/grok-imagine-video-1-5.md): $0.08/sec 480p, $0.14/sec 720p, $0.25/sec 1080p; $0.01/input image
- [P-Video-Edit](https://inference.sh/apps/pruna/p-video-edit.md): $0.025-$0.045/sec output video; draft: $0.025/sec, standard: $0.045/sec
- [Mirage Avatar X](https://inference.sh/apps/mirage/avatar-x.md): $0.175/sec
- [Seedance 2.5](https://inference.sh/apps/bytedance/seedance-2-5.md): $6.40–$11.70/M tokens, varies by output resolution and visual input
- [Seedance 2.5 Studio](https://inference.sh/apps/bytedance/seedance-2-5-studio.md): $6.40–$11.70/M tokens, varies by output resolution and visual input

## faq

### how much does MiniMax H3 Max cost?

$0.08/sec (768P), $0.05/sec (480P). you pay per run from your inference shell credits. no subscription is required.

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

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

### who makes MiniMax H3 Max?

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

### can my agent use MiniMax H3 Max?

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 Max?

other video apps on inference shell include Grok Imagine Video 1.5, P-Video-Edit, Mirage Avatar X, Seedance 2.5. they run through the same api, so switching is a one-line change.