Built-in Tools (Multimodal Models)

Multimodal Model Tools

Multimodal capabilities are provided by a combination of built-in backend tools and external tool services.

Common providers and modules live in monkey-tools-third-party-api, including image, video, visual-generation, media, and provider-specific adapters such as Fal, Volcengine Visual, Runway, Tripo, Google Gemini, Vertex AI, and OpenAI.

Usage Pattern

1. Deploy the tool service.
2. Import its manifest.json into Studio.
3. Add the corresponding tool node to a workflow or agent.
4. Store generated files in the configured object storage or asset system.