GPT-5.4 Pro: benchmarks, price and rank
GPT-5.4 Pro from OpenAI is 19th of 123 rated AI models on Spezia's leaderboard, strongest on GPQA Diamond. It costs $30 in, $180 out per million tokens.
- 19th of 123 rated models
- Rating 1,600
- Rated on 6 benchmarks
- Closed
- Released Mar 5, 2026
- 1.1 million tokens of context
Where it is strongest
- 8th of 103 on GPQA Diamond
Where it falls behind
In the top half on every benchmark it has a result on.
Price: $30 in, $180 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.
- Humanity's Last Exam44.3%
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.
- SimpleBench74.1%
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.
19th of 68. Independent result from Epoch AI.
- SimpleQA Verified46.3%
Short factual questions with one checked answer, asked without web search.
39th of 78. Independent result from Epoch AI.
- FrontierMath82.5%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
17th of 76. Independent result from Epoch AI.
- GPQA Diamond94.6%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
8th of 103. Independent result from Epoch AI.
- ARC-AGI-283.3%
Visual puzzles of colored grids, each with a new rule to work out from a few examples.
18th of 69. Independent result from Epoch AI.
- FrontierMath Tier 458.5%
Research-level problems that take professional mathematicians days.
21st of 66. Independent result from Epoch AI.
- Epoch Capabilities Index160.2
One score built from dozens of benchmarks, set on a common scale so that models tested on different benchmarks can still be compared.
19th of 94. Independent result from Epoch AI.
Its full profile is in Spezia.
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.