Scores updated · 14 tests both models report · How we compare
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.
Other results14 tests
- MATHGemma 2 Instruct (9B) by 59.372.413.1+59.3
- Social IQAMistral 7B by 21.253.474.6+21.2
- AGI EVALGemma 2 Instruct (9B) by 17.752.835.1+17.7
- GSM8KGemma 2 Instruct (9B) by 16.468.652.2+16.4
- BoolQGemma 2 Instruct (9B) by 1284.272.2+12
- OpenBookQAGemma 2 Instruct (9B) by 11.29179.8+11.2
- HumanEvalGemma 2 Instruct (9B) by 10.940.229.3+10.9
- ARC-ChallengeGemma 2 Instruct (9B) by 8.468.460+8.4
- PIQAGemma 2 Instruct (9B) by 481.777.7+4
- WinoGrandeGemma 2 Instruct (9B) by 2.280.678.4+2.2
- TriviaQAGemma 2 Instruct (9B) by 1.476.675.2+1.4
- HellaSwagMistral 7B by 1.381.983.2+1.3
- MBPPGemma 2 Instruct (9B) by 1.352.451.1+1.3
- MMLUtie64.564.2tie
Questions people ask
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. None of them is in the eight capability areas we count, so this page lists them without a verdict. Each score is the one shown on the model's own page.