---
title: "o3 mini high Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for o3 mini high."
source: "https://rankedagi.com/models/o3-mini-high"
---

# o3 mini high

| Field | Value |
| --- | --- |
| Organization | OpenAI |
| License | Proprietary |
| Version | Latest |
| Released | 2025-01-31 |
| Context window | 200K |
| Knowledge cutoff | 2023-10-01 |
| Input cost per million tokens | $1.1 |
| Output cost per million tokens | $4.4 |
| Cached input cost per million tokens | $0.55 |
| Model ID | o3-mini |
| Last updated | 2026-05-04T00:13:11.660Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 51.9% | RankedAGI Coding Score |  |
| SWEBench Verified | coding | 49.3% | Agentic Coding | https://www.swebench.com/index.html |
| RankedAGI Agentic | Agents | 40.8% | RankedAGI Agentic Score |  |
| LiveCodeBench v6 | coding | 68.9% | LiveCodeBench v6 - Evaluation of Large Language Models for Code |  |
| LiveCodeBench v5 | coding | 80.5% | LiveCodeBench - Evaluation of Large Language Models for Code |  |
| ChatArena (LMSYS) | coding | 1332 | ChatArena (LMSYS) Coding ELO Score with Style Control |  |
| Codeforces ELO Score | coding | 2130 | Codeforces ELO Score |  |
| Aider Polyglot | coding | 60.4% | Aider Polyglot Code Completion Benchmark |  |
| LiveBench Coding Score (old) | coding | 82.7% | LiveBench Coding Score old version |  |
| RankedAGI Reasoning | reasoning | 50.6% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 14.0% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 79.7% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1363 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| AIME 2025 | math | 86.5% | AIME 2025 Competition Math |  |
| AIME 2024 | math | 87.3% | AIME 2024 Competition Math |  |
| NYT Connections Extended | reasoning | 61.4% | NYT Connections Extended Version | https://github.com/lechmazur/nyt-connections/ |
| MMLU | knowledge | 86.9% | Image Reasoning |  |
| Hughes Hallucination Evaluation | safety | 0.8% | 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 | 75.8% | LiveBench Average Score old version |  |
| LiveBench Coding 25.4 | coding | 65.5% | LiveBench Coding Score 2025-04 | https://livebench.ai/ |
| LiveBench Data Analysis | general | 70.6% | LiveBench Data Analysis Score | https://livebench.ai/ |
| LiveBench Language | general | 50.7% | LiveBench Language Score | https://livebench.ai/ |
| RankedAGI Math | math | 68.7% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 49.9% | Overall RankedAGI score |  |
| GDPval-AA Elo | Agents | 748 | Office Tasks (Artificial Analysis) |  |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://openai.com/index/openai-o3-mini/ |