---
title: "GPT 5.2 Pro Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for GPT 5.2 Pro."
source: "https://rankedagi.com/models/gpt-5-2-pro"
---

# GPT 5.2 Pro

| Field | Value |
| --- | --- |
| Organization | OpenAI |
| License | Proprietary |
| Type | Thinking |
| Version | Latest |
| Released | 2025-12-11 |
| Knowledge cutoff | 2025-08-31 |
| Input cost per million tokens | $21 |
| Output cost per million tokens | $168 |
| Last updated | 2026-04-19T18:20:06.627Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 73.1% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 69.8% | RankedAGI Agentic Score |  |
| BrowseComp | Agents | 77.9% | A benchmark for browsing agents | https://openai.com/index/browsecomp/ |
| RankedAGI Reasoning | reasoning | 70.4% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 36.6% | Multidisciplinary Reasoning (no tools) |  |
| Humanity's Last Exam w/ Tools | reasoning | 50.0% | Multidisciplinary Reasoning (with tools) |  |
| GPQA Diamond | reasoning | 93.2% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| AIME 2025 | math | 100.0% | AIME 2025 Competition Math |  |
| RankedAGI Math | math | 76.2% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 65.3% | Overall RankedAGI score |  |
| ARC AGI 2 | reasoning | 54.2% | Abstract Reasoning Puzzles (Public) |  |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://openai.com/index/introducing-gpt-5-2/ |