Scores updated · 17 tests both models report · How we compare
Every test, side by side
All 17 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results17 tests
- SuperGPQAGLM-4.5-Base by 8.649.641+8.6
- GPQA (unspecified)Step3.5 Flash Base by 8.233.541.7+8.2
- MBPP+ (EvalPlus-augmented)GLM-4.5-Base by 7.57870.6+7.5
- WinoGrandeGLM-4.5-Base by 6.185.279.1+6.1
- MATHStep3.5 Flash Base by 5.86166.8+5.8
- Chinese SimpleQA (C-SimpleQA)GLM-4.5-Base by 5.368.563.2+5.3
- C-EvalStep3.5 Flash Base by 3.885.889.6+3.8
- HumanEvalStep3.5 Flash Base by 2.978.281.1+2.9
- MMLU-ReduxStep3.5 Flash Base by 2.686.689.2+2.6
- CMMLUStep3.5 Flash Base by 2.486.588.9+2.4
- SimpleQAStep3.5 Flash Base by 2.329.331.6+2.3
- MBPPGLM-4.5-Base by 2.281.679.4+2.2
- BBHStep3.5 Flash Base by 286.288.2+2
- GSM8KGLM-4.5-Base by 1.990.188.2+1.9
- MMLUGLM-4.5-Base by 1.987.785.8+1.9
- MMLU-ProGLM-4.5-Base by 1.463.762.3+1.4
- HellaSwagtie90.290.2tie
Questions people ask
How do you compare the two?
We use the 17 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.