io.github.AceDataCloud/mcp-minimax
MCP server for MiniMax H3 multimodal video generation
Open source Repository Open in the app JSON README (API)
About
MCP server for MiniMax H3 multimodal video generation
Details
- Kind
- MCP servers
- Topic
- Media, design & games
- Publisher
- acedatacloud
- Origin
- official
- Category
- ferramentas
- Transport
- http
- Version
- 2026.8.28.0
- Stars
- 1
- Open pull requests
- 7
- Last push
- 2026-08-28T10:41:15Z
- Repository state
- ativo
- Language
- Python
- License
- MIT
- Added
- 2026-08-29 03:01:40
- Updated
- 2026-08-29 03:01:40
- Origin id
io.github.AceDataCloud/mcp-minimax
README
# MiniMax H3 MCP
<!-- mcp-name: io.github.AceDataCloud/mcp-minimax -->
[](https://pypi.org/project/mcp-minimax/)
[](https://python.org)
[](LICENSE)
Model Context Protocol server for MiniMax H3 multimodal video generation through AceDataCloud.
## Capabilities
- Text-to-video
- Image-, video-, and audio-guided video generation
- 4–15 second output in 768P or 2K, with adaptive, landscape, portrait, and square ratios
- Optional asynchronous submission, task retrieval, batch retrieval, and deletion
- Hosted OAuth HTTP transport and local stdio transport
## Tools
| Tool | Purpose |
| --- | --- |
| `minimax_generate_video_from_text` | Generate from a detailed prompt |
| `minimax_generate_video_from_images` | Generate from one to nine image URLs |
| `minimax_generate_video_from_audio` | Generate with one to three audio URLs and required images |
| `minimax_generate_video` | Generate from the full documented content schema |
| `minimax_list_tasks` | List tasks with optional timestamp filters |
| `minimax_get_task` | Retrieve one task |
| `minimax_get_tasks_batch` | Retrieve several tasks |
| `minimax_delete_task` | Delete one task |
| `minimax_list_models` | Show model constraints |
| `minimax_list_actions` | Show available workflows |
## Hosted server
```text
https://minimax.mcp.acedata.cloud/mcp
```
The hosted server supports AceDataCloud OAuth. MCP clients that support remote OAuth can connect directly to this URL.
Public API reference: [MiniMax H3 Videos API](https://platform.acedata.cloud/documents/minimax-videos).
## Local installation
```bash
pipx install mcp-minimax
export ACEDATACLOUD_API_TOKEN="YOUR_API_TOKEN"
mcp-minimax
```
Get a token from [AceDataCloud](https://platform.acedata.cloud/console/applications).
### Claude Code
```bash
claude mcp add minimax --transport stdio \
--env ACEDATACLOUD_API_TOKEN=YOUR_API_TOKEN \
-- mcp-minimax
```
### Generic MCP configuration
```json
{
"mcpServers": {
"minimax": {
"command": "mcp-minimax",
"env": {
"ACEDATACLOUD_API_TOKEN": "YOUR_API_TOKEN"
}
}
}
}
```
## Examples
Text:
```text
Generate a 6-second 16:9 video: a red fox running through a snowy forest at dawn, low tracking shot.
```
Images:
```text
Animate these two reference images for 8 seconds while preserving the character and clothing.
```
Audio:
```text
Create a 9:16 dance video guided by this audio, with cuts and motion following the beat.
```
The HTTP API waits for completion by default, while MCP generation tools default `async` to `true` so agents receive a task ID immediately. Poll it with `minimax_get_task` until the final AceDataCloud CDN video is available. Set `async=false` only when the client can safely wait for the complete result.
## Development
```bash
pip install -e ".[dev,test]"
ruff check .
pytest --cov=core --cov=tools
mypy core tools
```
## Documentation
<!-- canonical-documentation -->
[Documentation](https://platform.acedata.cloud/documents/minimax-videos)
## License
MIT