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

# GPT-5

| Field | Value |
| --- | --- |
| Organization | OpenAI |
| License | Proprietary |
| Type | Non-thinking |
| Version | Latest |
| Released | 2025-08-07 |
| Context window | 400K |
| Knowledge cutoff | 2024-10-01 |
| Input cost per million tokens | $1.25 |
| Output cost per million tokens | $10 |
| Cached input cost per million tokens | $0.13 |
| Model ID | gpt-5-2025-08-07 |
| Last updated | 2026-05-04T03:43:08.354Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 54.4% | RankedAGI Coding Score |  |
| SWEBench Pro | coding | 36.3% | Diverse Agentic Coding |  |
| SWEBench Verified | coding | 52.8% | Agentic Coding | https://www.swebench.com/index.html |
| RankedAGI Agentic | Agents | 50.1% | RankedAGI Agentic Score |  |
| Code Arena | coding | 1393 | Code Arena Elo Score | https://arena.ai/leaderboard/code |
| SvelteBench | coding | 85.6% | SvelteBench - Benchmark for Svelte | https://khromov.github.io/svelte-bench/benchmark-results-merged.html |
| CyberGym | safety | 60.2% |  |  |
| Aider Polyglot | coding | 26.7% | Aider Polyglot Code Completion Benchmark |  |
| RankedAGI Reasoning | reasoning | 45.7% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 6.3% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 77.8% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1426 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| AIME 2025 | math | 94.6% | AIME 2025 Competition Math |  |
| MMMU | imaging | 74.4% | Multimodal Understanding College-level visual problem-solving |  |
| MMMU-Pro | imaging | 62.7% | Multimodal Understanding |  |
| RankedAGI Math | math | 67.6% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 51.5% | Overall RankedAGI score |  |
| 𝜏²-Bench Telecom | Agents | 38.6% | Agentic Tool Use |  |
| 𝜏²-Bench Retail | Agents | 72.8% | Agentic Tool Use |  |
| 𝜏²-Bench Airline | Agents | 55.0% | Agentic Tool Use |  |
| HealthBench Hard | Health | 25.5% | Open-ended Health Queries |  |
| GDPval-AA Elo | Agents | 392 | Office Tasks (Artificial Analysis) |  |
| SvelteBench old | coding | 78.9% | SvelteBench - Benchmark for Svelte (Old v1) | https://khromov.github.io/svelte-bench/v1/v1-benchmark-results-merged.html |
| Code DesignArena | Design | 1215 |  | https://www.designarena.ai/leaderboard/code |
| Toolathlon | Agents | 37.7% | 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/ |