Model card
o1-preview.
OpenAIcommercialReasoning LLM2 current SOTA
OpenAI's reasoning-focused model.
§ 02 · Benchmarks
Every benchmark o1-preview has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | HumanEval | Code & Software Engineering · Code Generation | pass@1 | 92.4% | #8 | 2026-04-20 | source ↗ |
| 02 | MATH | Language & Knowledge · Mathematical Reasoning | accuracy | 85.5% | #19 | 2026-04-20 | |
| 03 | GSM8K | Language & Knowledge · Mathematical Reasoning | accuracy | 97.8% | #1 | 2026-04-20 | |
| 04 | MMLU | — | accuracy | 90.8% | #3 | 2026-04-20 | |
| 05 | GPQA | — | accuracy | 73.3% | #5 | 2026-04-20 | |
| 06 | AIME 2024 | Language & Knowledge · Mathematical Reasoning | accuracy | 83.3% | #1 | 2026-04-20 |
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 o1-preview actually performs.
§ 06 · Sources & freshness
Where these numbers come from.
src
5
results
codesota-api
1
result
0 of 6 rows marked verified.