Key Features

What you can do

Integrated Tool Use in Chain-of-Thought

Supports web browsing, Python code execution, image and file analysis, image generation, canvas, automations, file search, and memory within its reasoning process.

Visual Reasoning with Image Transformations

Processes images natively with cropping, zooming, and rotating capabilities during reasoning without separate models.

Adjustable Reasoning Effort and Debugging

Allows users to set reasoning effort levels (low, medium, high) and access full chain-of-thought for debugging purposes.

Large Context Window and Multi-Modal Input

Supports a 200,000 token context window and accepts both text and image inputs with text outputs.

Function Calling and Structured Outputs

Enables function calling and structured output formats to facilitate integration and automation.