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

# GPT-5 mini

| Field | Value |
| --- | --- |
| Organization | OpenAI |
| License | Proprietary |
| Version | Latest |
| Released | 2025-08-07 |
| Context window | 400K |
| Knowledge cutoff | 2024-05-31 |
| Input cost per million tokens | $0.25 |
| Output cost per million tokens | $2 |
| Cached input cost per million tokens | $0.03 |
| Model ID | gpt-5-mini-2025-08-07 |
| Last updated | 2026-05-04T00:11:22.887Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 50.8% | RankedAGI Coding Score |  |
| SWEBench Pro | coding | 45.7% | Diverse Agentic Coding |  |
| SWEBench Verified | coding | 71.0% | Agentic Coding | https://www.swebench.com/index.html |
| Terminal-Bench 2.0 | coding | 38.2% | Agentic Terminal Coding | https://www.tbench.ai/leaderboard/terminal-bench/2.0 |
| RankedAGI Agentic | Agents | 41.9% | RankedAGI Agentic Score |  |
| Livebench Coding 26.1 | coding | 76.1% | Livebench Coding 26.01.08 | https://livebench.ai/#/?Coding=a&sort=Coding+Average&variants=true&highunseenbias=true |
| OSWorld-Verified | Agents | 42.0% |  |  |
| SvelteBench | coding | 23.3% | SvelteBench - Benchmark for Svelte | https://khromov.github.io/svelte-bench/benchmark-results-merged.html |
| Aider Polyglot | coding | 71.6% | Aider Polyglot Code Completion Benchmark |  |
| RankedAGI Reasoning | reasoning | 54.7% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 18.3% | Multidisciplinary Reasoning (no tools) |  |
| Humanity's Last Exam w/ Tools | reasoning | 18.3% | Multidisciplinary Reasoning (with tools) |  |
| GPQA Diamond | reasoning | 82.3% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1390 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| AIME 2025 | math | 91.1% | AIME 2025 Competition Math |  |
| Vending Bench 2 | Agents | $-31.18 | Benchmark for measuring AI model performance on running a business over long time horizons. Models are tasked with running a simulated vending machine business over a year and scored on their bank account balance at the end. | https://andonlabs.com/evals/vending-bench-2 |
| MMMU-Pro | imaging | 67.5% | Multimodal Understanding |  |
| MMMU Pro w/Tools | imaging | 74.1% |  |  |
| RankedAGI Math | math | 72.0% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 51.1% | Overall RankedAGI score |  |
| 𝜏²-Bench Telecom | Agents | 74.1% | Agentic Tool Use |  |
| GDPval-AA Elo | Agents | 1190 | Office Tasks (Artificial Analysis) |  |
| MCP Atlas | Agents | 47.6% | Scaled Tool Use |  |
| SvelteBench old | coding | 21.1% | SvelteBench - Benchmark for Svelte (Old v1) | https://khromov.github.io/svelte-bench/v1/v1-benchmark-results-merged.html |
| Code DesignArena | Design | 1165 |  | https://www.designarena.ai/leaderboard/code |
| Toolathlon | Agents | 14.5% | Benchmark to assess language agents' general tool use in realistic environments. | https://toolathlon.xyz/introduction |

## Model Links

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