Gemini 2.0 Pro

// Overview

Gemini 2.0 Pro was Google's experimental flagship in the Gemini 2.0 family, announced on February 5, 2025 as part of a wave that also expanded Gemini 2.0 Flash to general availability and introduced Gemini 2.0 Flash-Lite. Google positioned Gemini 2.0 Pro (shipped as the model id gemini-2.0-pro-exp-02-05) as its best model yet for coding performance and complex prompts, citing stronger understanding and reasoning over world knowledge than any Gemini it had released to that point.

Its headline feature was scale of context: Gemini 2.0 Pro launched with Google's largest context window at the time, 2 million tokens, letting it ingest very large documents, transcripts, and codebases in a single prompt. It accepted text and image input and produced text output, and could call tools such as Google Search and code execution. The model was released to developers in Google AI Studio and Vertex AI, and to Gemini Advanced subscribers via the in-app model picker. Because it was an experimental release, it was offered for free with no published paid pricing.

Gemini 2.0 Pro was deliberately short-lived. As an experimental preview it was always subject to change, and Google's thinking-native Gemini 2.5 Pro arrived just weeks later on March 25, 2025 and effectively replaced it. Google formally deprecated gemini-2.0-pro-exp-02-05 on November 4, 2025 and shut it down on December 2, 2025, so the model is no longer available through Google's APIs.

// Benchmarks

1. MMLU-Pro (reasoning & knowledge)79.1%
2. GPQA Diamond (science)64.7%
3. MMMU (multimodal)72.7%
4. AIME 2025 (math)27.5%

Scores on a 0–100 scale (25-point gridlines); higher is better. Each benchmark links to its published source.

// Pricing

Gemini 2.0 Pro was an experimental model and was offered free of charge; Google never published a paid per-token price for it before its December 2025 shutdown.

Pricing source ↗

// Strengths

• Strongest coding and complex-prompt performance of the Gemini 2.0 generation, per Google's launch positioning
• 2-million-token context window — the largest Gemini offered at launch — for reasoning over very large documents and codebases
• Built-in tool use: Google Search grounding, code execution, and function calling
• Multimodal text-plus-image input for analyzing documents, charts, and screenshots
• Free to use during its experimental phase across Google AI Studio, Vertex AI, and Gemini Advanced

// Best for

• Coding assistance and complex multi-step prompts (its primary positioning)
• Long-context tasks: reasoning over large documents, transcripts, and code repositories in one pass
• Agentic workflows using Google Search grounding and code execution
• Multimodal analysis of text combined with images
• Prototyping and experimentation in Google AI Studio or Vertex AI before moving to a stable model

// How to access

// Gemini Pro — every version

The full lineage of the Gemini Pro line, newest first. Every version has its own page — click any to compare specs, benchmarks and pricing.