Scores updated · 13 tests both models report · How we compare
Every test, side by side
All 13 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results13 tests
- SimpleQAKimi K2 Base by 27.47.935.3+27.4
- MBPPKimi K2 Base by 25.263.889+25.2
- MMLU-ProKimi K2 Base by 24.644.669.2+24.6
- HumanEval+Kimi K2 Base by 2262.884.8+22
- MGSMKimi K2 Base by 21.563.785.2+21.5
- MBPP+ (EvalPlus-augmented)Kimi K2 Base by 19.85473.8+19.8
- GSM8KKimi K2 Base by 18.673.592.1+18.6
- IncludeKimi K2 Base by 18.157.675.7+18.1
- BBHKimi K2 Base by 14.973.888.7+14.9
- MMLUKimi K2 Base by 14.273.687.8+14.2
- MMMLUKimi K2 Base by 12.964.777.6+12.9
- DROPKimi K2 Base by 11.272.483.6+11.2
- HumanEvalKimi K2 Base by 9.968.378.2+9.9
Questions people ask
How do you compare the two?
We use the 13 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.