GPT-5: benchmarks, price and rank
GPT-5 from OpenAI is 75th of 123 rated AI models on Spezia's leaderboard. It costs $1.25 in, $10 out per million tokens.
- 75th of 123 rated models
- Rating 1,038
- Rated on 12 benchmarks
- Closed
- Released Aug 7, 2025
- 400,000 tokens of context
- Knows the world to September 2024
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 15th of 17 on Remote Labor Index
- 54th of 69 on ARC-AGI-2
- 81st of 107 on LMArena text
Price: $1.25 in, $10 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- Remote Labor Index1.7%
Real freelance projects that people were paid to do, from design and video to architecture, games and data work. The model gets the same brief and files the freelancer got and has to hand back finished work.
15th of 17. Independent result from Epoch AI.
- BrowseComp54.9%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
40th of 56. The maker's own number, from OpenAI, via LLM Stats.
- Humanity's Last Exam25.3%
About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.
Independent result from Epoch AI.
- SimpleBench56.7%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
48th of 68. Independent result from Epoch AI.
- SimpleQA Verified50.1%
Short factual questions with one checked answer, asked without web search.
31st of 78. Independent result from Epoch AI.
- FrontierMath55.4%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
51st of 76. Independent result from Epoch AI.
- MMMU-Pro78.4%
College exam questions that need a picture to answer, such as charts, diagrams, maps and chemical structures, with more answer options than the original MMMU and some questions shown only as a screenshot.
20th of 51. The maker's own number, from OpenAI, via LLM Stats.
- CharXiv Reasoning81.1%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
25th of 45. The maker's own number, from OpenAI, via LLM Stats.
- LMArena text1,406
People put a question to two anonymous models and vote for the better answer.
81st of 107. Independent result from LMArena.
- LMArena WebDev1,417
People ask two anonymous models to build the same web page or app and vote for the better one.
67th of 98. Independent result from LMArena.
10 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.