Model card
GPT-4.5 Preview.
OpenAIapi
§ 02 · Benchmarks
Every benchmark GPT-4.5 Preview has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | MMLU | Reasoning · Commonsense Reasoning | accuracy | 90.8% | #8 | 2025-02-27 | source ↗ |
| 02 | HumanEval | Computer Code · Code Generation | pass@1 | 88.6% | #21 | — | source ↗ |
| 03 | MATH | Reasoning · Mathematical Reasoning | accuracy | 87.1% | #22 | — | source ↗ |
| 04 | GPQA Diamond | Reasoning · Multi-step Reasoning | accuracy | 69.5% | #52 | — | source ↗ |
| 05 | HLE | Reasoning · Multi-step Reasoning | accuracy | 5.4% | #65 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where GPT-4.5 Preview actually performs.
§ 05 · Related models
Other OpenAI models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
openai-simple-evals
4
results
scale-hle-official
1
result
2 of 5 rows marked verified.