Google18:58Feature UpdatesOfficial Docs
Gemini API Adds Multimodal Function Calls with Image Support
AI can handle tool results including images directly, streamlining agent development.
Key Points
- 1Supports mixed image and text function results
- 2Processes screenshots and other visuals
- 3Available in gemini-3-flash-preview
- 4Python usage guide released
Gemini Interactions API now supports multimodal function calls, letting tools return images alongside text. Gemini processes visual data, empowering agents to handle visual tasks more effectively.