Claude 3 Opus wins more areas, narrowly.
Scores updated · 10 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 Claude 3 Opus pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+18.7points ahead
- Code for real scientific research problemsSciCode+9.9points ahead
Where Mistral Small (2402) pulls ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+1.3points ahead
Every test, side by side
All 10 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningEven
- GPQA DiamondClaude 3 Opus by 18.748.930.2+18.7
- Humanity's Last ExamMistral Small (2402) by 1.32.84.1+1.3
Other results7 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.
- MMLUClaude 3 Opus by 27.586.859.3+27.5
- GSM8KClaude 3 Opus by 21.69573.4+21.6
- NarrativeQAMistral Small (2402) by 16.835.151.9+16.8
- naturalquestions_closedbookClaude 3 Opus by 13.74430.4+13.7
- OpenBookQAClaude 3 Opus by 9.495.686.2+9.4
- AA IntelligenceClaude 3 Opus by 3.28.75.5+3.2
- MATHMistral Small (2402) by 260.162+2
Questions people ask
Which is better, Claude 3 Opus or Mistral Small (2402)?
Claude 3 Opus wins one of the two areas where both have results: coding. Mistral Small (2402) wins none. They are level on reasoning.
Which is better for coding?
Claude 3 Opus. It wins the one coding test both models report.
How do you compare the two?
We use the 10 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 7 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.