---
title: "Gemini 2.5 Deep Think Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Gemini 2.5 Deep Think."
source: "https://rankedagi.com/models/gemini-2.5-deep-think"
---

# Gemini 2.5 Deep Think

| Field | Value |
| --- | --- |
| Organization | Google |
| License | Proprietary |
| Version | Latest |
| Released | 2025-08-01 |
| Last updated | 2025-08-08 01:09:28.891528+00 |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 71.9% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 54.3% | RankedAGI Agentic Score |  |
| LiveCodeBench v6 | coding | 87.6% | LiveCodeBench v6 - Evaluation of Large Language Models for Code |  |
| RankedAGI Reasoning | reasoning | 70.0% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 34.8% | Multidisciplinary Reasoning (no tools) |  |
| AIME 2025 | math | 99.2% | AIME 2025 Competition Math |  |
| RankedAGI Math | math | 75.6% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 61.6% | Overall RankedAGI score |  |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://blog.google/products/gemini/gemini-2-5-deep-think/ |