A Node.js Model Context Protocol (MCP) server that exposes Together AI's inference endpoints — chat completions, image generation, vision, and embeddings — as tools callable from Claude Desktop, Cursor, VS Code, and any other MCP-compatible client.
I created this MCP due to an issue I was having accessing reasoning models through Together AI.
Together AI's largest reasoning models (GLM-5, Qwen3.5-397B, MiniMax M2.5, Kimi K2.5) use a non-standard response format. During chain-of-thought generation, these models write their reasoning trace into choices[0].message.reasoning while leaving choices[0].message.content as an empty string. The final answer only appears in message.content once thinking is complete.
サーバー設定
{
"mcpServers": {
"together-ai": {
"command": "node",
"args": [
"/absolute/path/to/together-ai-mcp/index.js"
],
"env": {
"TOGETHER_API_KEY": "your_api_key_here",
"IMAGE_OUTPUT_DIR": "/path/to/save/images"
}
}
}
}