Scores updated · 15 tests both models report · How we compare
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.
Granite 4.1 3B BaseNo current price is tracked.
Granite 4.1 8B$0.05 to read · $0.10 to writePrice from ibm-granite
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results15 tests
- Minerva MathGranite 4.1 8B by 42.13880.1+42.1
- HumanEval+Granite 4.1 8B by 25.654.379.9+25.6
- MGSMGranite 4.1 8B by 23.858.582.3+23.8
- GSM8KGranite 4.1 8B by 19.672.992.5+19.6
- MMLU-ProGranite 4.1 8B by 18.837.256+18.8
- AGI EVALGranite 4.1 8B by 18.154.372.4+18.1
- BBHGranite 4.1 8B by 16.763.880.5+16.7
- Eval+ AvgGranite 4.1 8B by 14.365.980.2+14.3
- HumanEvalGranite 4.1 8B by 9.276.285.4+9.2
- MMMLUGranite 4.1 8B by 8.256.664.8+8.2
- MMLUGranite 4.1 8B by 7.366.573.8+7.3
- IncludeGranite 4.1 8B by 7.151.858.9+7.1
- MBPPGranite 4.1 8B by 5.881.587.3+5.8
- MBPP+ (EvalPlus-augmented)Granite 4.1 8B by 5.568.373.8+5.5
- SimpleQAGranite 4.1 3B Base by 2.16.84.8+2.1
Questions people ask
How do you compare the two?
We use the 15 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.