Claude 3 Sonnet is the stronger all-rounder.
Scores updated · 16 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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude 3 Sonnet pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+7.6points ahead
- Code for real scientific research problemsSciCode+4.8points ahead
Where Gemini 1.5 Flash (001) pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 16 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningClaude 3 Sonnet
- GPQA DiamondClaude 3 Sonnet by 7.64032.4+7.6
- Humanity's Last Examtie3.64.4tie
Other results13 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.
- NarrativeQAGemini 1.5 Flash (001) by 63.511.174.6+63.5
- GSM8KClaude 3 Sonnet by 59.592.332.8+59.5
- MATHGemini 1.5 Flash (001) by 47.743.190.8+47.7
- naturalquestions_closedbookGemini 1.5 Flash (001) by 29.52.832.3+29.5
- AIR-Bench 2024Claude 3 Sonnet by 17.684.767.1+17.6
- HarmBenchClaude 3 Sonnet by 15.895.880+15.8
- MMLUClaude 3 Sonnet by 10.478.367.9+10.4
- XSTestGemini 1.5 Flash (001) by 6.385.892.1+6.3
- bbqGemini 1.5 Flash (001) by 4.79094.7+4.7
- simple_safety_testsClaude 3 Sonnet by 310097+3
- OpenBookQAtie91.891.4tie
- anthropic_red_teamtie99.899.9tie
- AA Intelligencetie5.95.9tie
Questions people ask
Which is better, Claude 3 Sonnet or Gemini 1.5 Flash (001)?
Claude 3 Sonnet wins all two areas where both have results: coding and reasoning. Gemini 1.5 Flash (001) wins none.
Which is better for coding?
Claude 3 Sonnet. It wins the one coding test both models report.
How do you compare the two?
We use the 16 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 13 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.