Gemini 2.0 Flash Thinking is engineered to "think out loud," providing users with a transparent view of its problem-solving process. This approach not only enhances the model's reasoning capabilities but also allows users to follow and understand the logical steps leading to a solution.
Transformer-based model with reasoning-specific layers optimized for multimodal inputs (text/images). Includes advanced explainability mechanisms to break down reasoning steps.
Trained on diverse datasets including math/science benchmarks (e.g., AIME2024), multimodal corpora combining text and images, and synthetic reasoning datasets.
Incorporates billions of tokens from curated sources alongside millions of structured reasoning examples.
The model is available on the AI/ML API platform as "Gemini 2.0 Flash Thinking" .
Detailed API Documentation is available here.
Google DeepMind emphasizes ethical considerations in AI development by promoting transparency regarding the model's capabilities and limitations. The organization encourages responsible usage to prevent misuse or harmful applications of generated content.
Gemini 2.0 Flash Thinking is available under a commercial license that allows both research and commercial usage rights while ensuring compliance with ethical standards regarding creator rights.
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 8.9 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Math | 21.7 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Gemini 2.0 Flash Thinking This page | ||||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
Yes, Gemini 2.0 Flash Thinking can stream responses as they are generated.
Yes, Gemini 2.0 Flash Thinking supports both function calling and structured outputs.
Gemini 2.0 Flash Thinking was built by Google.
Its listed capabilities are function calling, streaming, structured outputs, and reasoning; image generation is not included.
It is built for reasoning tasks and also supports function calling and structured output generation, making it useful for tasks needing logical steps and tool integration.
It is available through the v1 endpoint.
