Scores updated · 16 tests both models report · How we compare
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.
Other results16 tests
- BBHGranite 4.1 30B Base by 16.980.763.8+16.9
- MMMLUGranite 4.1 30B Base by 16.873.456.6+16.8
- MGSMGranite 4.1 30B Base by 15.974.458.5+15.9
- IncludeGranite 4.1 30B Base by 15.367.151.8+15.3
- AGI EVALGranite 4.1 30B Base by 14.969.254.3+14.9
- DROPGranite 4.1 30B Base by 12.678.666+12.6
- MMLU-ProGranite 4.1 30B Base by 12.349.537.2+12.3
- MMLUGranite 4.1 30B Base by 11.978.466.5+11.9
- GSM8KGranite 4.1 30B Base by 10.983.872.9+10.9
- HumanEvalGranite 4.1 3B Base by 8.567.776.2+8.5
- HumanEval+Granite 4.1 30B Base by 7.962.254.3+7.9
- Minerva MathGranite 4.1 30B Base by 7.745.738+7.7
- Eval+ AvgGranite 4.1 30B Base by 4.970.865.9+4.9
- SimpleQAGranite 4.1 30B Base by 3.610.56.8+3.6
- MBPPGranite 4.1 30B Base by 2.183.681.5+2.1
- MBPP+ (EvalPlus-augmented)Granite 4.1 30B Base by 1.369.668.3+1.3
Questions people ask
How do you compare the two?
We use the 16 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.