o3 is the stronger all-rounder.
Scores updated · 14 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Coding rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where o3 pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+20.5points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+13.9points ahead
- Code for real scientific research problemsSciCode+9.7points ahead
Where Gemini 2.0 Pro Exp 02.05 pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Reasoningo3
- GPQA Diamondo3 by 20.562.282.7+20.5
- Humanity's Last Examo3 by 13.96.220.1+13.9
Other results11 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- HMMT Feb. 2025o3 by 707.577.5+70
- AIME 2025o3 by 58.927.586.4+58.9
- Aider-Polygloto3 by 45.735.681.3+45.7
- HarmBencho3 by 33.165.398.4+33.1
- AIR-Bench 2024o3 by 16.168.484.5+16.1
- Artificial Analysis Coding Indexo3 by 12.925.538.4+12.9
- AA Intelligenceo3 by 11.58.720.2+11.5
- bbqo3 by 2.395.697.9+2.3
- XSTesto3 by 1.995.497.3+1.9
- simple_safety_testso3 by 1.597.599+1.5
- anthropic_red_teamtie98.798.3tie
Questions people ask
Which is better, Gemini 2.0 Pro Exp 02.05 or o3?
o3 wins all two areas where both have results: coding and reasoning. Gemini 2.0 Pro Exp 02.05 wins none.
Which is better for coding?
o3. It wins the one coding test both models report.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 3 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 11 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.