Scores updated · 11 tests both models report · How we compare
Every test, side by side
All 11 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results11 tests
- HumanEvalKimi K2 Base by 36.178.242.1+36.1
- MMLU-ProKimi K2 Base by 34.669.234.6+34.6
- TriviaQAKimi K2 Base by 3485.151.1+34
- BBHKimi K2 Base by 32.488.756.3+32.4
- MBPPKimi K2 Base by 31.98957.1+31.9
- MATHKimi K2 Base by 27.670.242.6+27.6
- MMLUKimi K2 Base by 22.287.865.6+22.2
- C-EvalKimi K2 Base by 17.592.575+17.5
- CMMLUKimi K2 Base by 15.990.975+15.9
- GSM8KKimi K2 Base by 1392.179.1+13
- CMathKimi K2 Base by 10.890.880+10.8
Questions people ask
How do you compare the two?
We use the 11 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.