Together AI Image MCP Server
# Together AI Image Server
An MCP server for generating images using Together AI's image generation models.
<a href="https://glama.ai/mcp/servers/2cwphjnpgw"><img width="380" height="200" src="https://glama.ai/mcp/servers/2cwphjnpgw/badge" alt="Together AI Image Server MCP server" /></a>
## Setup
1. Install dependencies:
```bash
npm install
```
2. Build the server:
```bash
npm run build
```
## Configuration
### 1. Together AI API Key
The server requires a Together AI API key. You can get one from [Together AI's platform](https://api.together.xyz/).
### 2. Cline Configuration
Add the server to your Cline MCP settings file:
For macOS/Linux: `~/Library/Application Support/Windsurf/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json`
```json
{
"mcpServers": {
"togetherai-image": {
"command": "node",
"args": ["/path/to/togetherai-image-server/build/index.js"],
"env": {
"TOGETHER_API_KEY": "your-api-key-here"
},
"disabled": false,
"autoApprove": []
}
}
}
```
## Usage in Cline
The server provides a `generate_image` tool with the following parameters:
```typescript
{
prompt: string; // Required: Text description of the image to generate
model?: string; // Optional: Model to use (default: 'black-forest-labs/FLUX.1.1-pro')
width?: number; // Optional: Image width in pixels (default: 1024)
height?: number; // Optional: Image height in pixels (default: 768)
steps?: number; // Optional: Number of inference steps (default: 28)
n?: number; // Optional: Number of images to generate (default: 1)
outputDir?: string; // Optional: Directory where images will be saved (default: './output')
}
```
### Example Usage
You can use the server in Cline like this:
```
Generate an image of a cat and save it to my desktop:
<use_mcp_tool>
<server_name>togeherai-image</server_name>
<tool_name>generate_image</tool_name>
<arguments>
{
"prompt": "A cute cat sitting on a windowsill",
"outputDir": "/Users/username/Desktop/generated-images"
}
</arguments>
</use_mcp_tool>
```
### Output
The tool returns a JSON response containing:
- `filepath`: Full path to the saved image
- `filename`: Name of the saved image file
- `dimensions`: Original and final dimensions of the image
## Features
- Supports Together AI's latest FLUX.1.1-pro model
- Automatically resizes images if dimensions are below 256 pixels
- Maintains aspect ratio during resizing
- High-quality JPEG output (90% quality)
- Creates output directories recursively
- Returns detailed metadata including image dimensions
TDQS
Scored across 1 tool
With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool 'generate_image' has a clear and distinct purpose, making it impossible for an agent to misselect between non-existent alternatives.
The naming pattern cannot be inconsistent with only one tool. The tool name 'generate_image' follows a verb_noun convention, which is straightforward and predictable, though there are no other tools to compare it against for consistency.
A single tool for an image generation server feels thin and under-scoped. While it covers the core functionality, typical image generation APIs offer more operations (e.g., variations, edits, or style transfers), making this count borderline too few for the apparent domain.
The tool set is severely incomplete for an image generation domain. It only provides generation, with no obvious gaps to assess since there are no other tools, but the lack of common operations like editing, upscaling, or batch processing suggests significant coverage issues for agent workflows.