o3 vs o4-mini
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | o3 | o4-mini |
|---|---|---|
| Coding | 8.0 o3 | 7.7 o4-mini |
| Agents | 8.6 o3 | 23.7* o4-mini |
| Reasoning | 38.9 o3 | 27.2 o4-mini |
| Knowledge & language | 49.3 o3 | 19.6 o4-mini |
| Human preference | 48.6 o3 | 46.7* o4-mini |
* Estimated: no published scores in that category. How the Respan Index works
Benchmarks both report
o3 leads on 7 of the 11 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | o3 | o4-mini |
|---|---|---|
| Chess Puzzles (Epoch AI) | 37% | 22% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 33.3% | 36.1% |
| GPQA Diamond | 81.8% | 77.6% |
| MATH Level 5 (Epoch AI) | 97.8% | 97.8% |
| Mystery Game Puzzles (Epoch AI) | 29% | 5% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 84.4% | 81.7% |
| SimpleBench (SimpleBench) | 53.1% | 38.7% |
| SimpleQA Verified | 49.4% | 19.6% |
| WeirdML (Håvard Tveit Ihle) | 52.4% | 52.6% |
| AIME 2025 | 89.2% | 91.7% |
| GSO Opt@1 (GSO) | 3.9% | 3.6% |
FAQ
Is o3 better than o4-mini?
o3 scores 26.1 on the Respan Index (#97), ahead of o4-mini at 23.6 (#99). o3 leads on 7 of the 11 benchmarks both report.
Which is cheaper, o3 or o4-mini?
o3 lists at $2 input and $8 output per 1M tokens; o4-mini at $1.1 and $4.4. o4-mini is 1.8x cheaper per token at list price.
Which has the larger context window?
Both accept 200K tokens of context.