Gemini 3 Flash adds active vision with Python code execution, lifting accuracy by 5 to 10%, so you can trust verified results.

What's new? Agentic Vision in Gemini 3 Flash uses a think act observe loop with Python code for visual analysis; available via Gemini API in Google AI Studio and Vertex AI; ...

Google DeepMind has added Agentic Vision to Gemini 3 Flash, enabling active image exploration through Python code execution with 5-10% quality improvements.

Agentic Vision is a new capability for Gemini 3 Flash to make image-related tasks more accurate by “grounding answers in visual evidence.” ...