---
title: "Gemini 2.5 Pro Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Gemini 2.5 Pro."
source: "https://rankedagi.com/models/gemini-2.5-pro-preview-03-25"
---

# Gemini 2.5 Pro

| Field | Value |
| --- | --- |
| Organization | Google |
| License | Proprietary |
| Version | Old |
| Released | 2025-03-25 |
| Context window | 1M |
| Knowledge cutoff | 2025-01-01 |
| Input cost per million tokens | $1.25 |
| Output cost per million tokens | $10 |
| Model ID | gemini-2.5-pro-preview-03-25 |
| Last updated | 2026-04-27T17:00:41.242Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 58.0% | RankedAGI Coding Score |  |
| SWEBench Verified | coding | 63.8% | Agentic Coding | https://www.swebench.com/index.html |
| RankedAGI Agentic | Agents | 49.0% | RankedAGI Agentic Score |  |
| Code Arena | coding | 1203 | Code Arena Elo Score | https://arena.ai/leaderboard/code |
| LiveCodeBench v5 | coding | 70.4% | LiveCodeBench - Evaluation of Large Language Models for Code |  |
| ChatArena (LMSYS) | coding | 1352 | ChatArena (LMSYS) Coding ELO Score with Style Control |  |
| Aider Polyglot | coding | 72.9% | Aider Polyglot Code Completion Benchmark |  |
| LiveBench Coding Score (old) | coding | 85.9% | LiveBench Coding Score old version |  |
| RankedAGI Reasoning | reasoning | 56.0% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 18.8% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 84.0% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1443 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| AIME 2025 | math | 86.7% | AIME 2025 Competition Math |  |
| AIME 2024 | math | 92.0% | AIME 2024 Competition Math |  |
| NYT Connections Extended | reasoning | 54.1% | NYT Connections Extended Version | https://github.com/lechmazur/nyt-connections/ |
| MMLU | knowledge | 89.8% | Image Reasoning |  |
| MMMU | imaging | 81.7% | Multimodal Understanding College-level visual problem-solving |  |
| Hughes Hallucination Evaluation | safety | 1.1% | Computed using Vectara's Hallucination Evaluation Model, also known as HHEM. This evaluates how often an LLM introduces hallucinations when summarizing a document. | https://github.com/vectara/hallucination-leaderboard |
| LiveBench Average | general | 82.3% | LiveBench Average Score old version |  |
| LiveBench Coding 25.4 | coding | 58.1% | LiveBench Coding Score 2025-04 | https://livebench.ai/ |
| LiveBench Data Analysis | general | 79.9% | LiveBench Data Analysis Score | https://livebench.ai/ |
| LiveBench Language | general | 67.8% | LiveBench Language Score | https://livebench.ai/ |
| RankedAGI Math | math | 71.7% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 54.5% | Overall RankedAGI score |  |
| SvelteBench old | coding | 83.3% | SvelteBench - Benchmark for Svelte (Old v1) | https://khromov.github.io/svelte-bench/v1/v1-benchmark-results-merged.html |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://blog.google/technology/google-deepmind/gemini-model-thinking-updates-march-2025/ |