The two are evenly matched.GLM-4.5-Base is better at coding; Kimi K2 Base at reasoning.
Scores updated · 35 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Coding and reasoning rest on a single test each.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where GLM-4.5-Base pulls ahead
- Recent programming contest problemsLiveCodeBench v6+1.1points ahead
Where Kimi K2 Base pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+13.2points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results33 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- RULER 128KKimi K2 Base by 88.6088.6+88.6
- RULER 64KKimi K2 Base by 77.716.193.8+77.7
- GPQA (unspecified)Kimi K2 Base by 9.633.543.1+9.6
- MATHKimi K2 Base by 9.26170.2+9.2
- MBPPKimi K2 Base by 7.481.689+7.4
- C-EvalKimi K2 Base by 6.785.892.5+6.7
- Chinese SimpleQA (C-SimpleQA)Kimi K2 Base by 6.168.574.6+6.1
- SimpleQAKimi K2 Base by 629.335.3+6
- MMLU-ProKimi K2 Base by 5.563.769.2+5.5
- SuperGPQAGLM-4.5-Base by 4.949.644.7+4.9
- MBPP-SanitizedGLM-4.5-Base by 4.676.772.1+4.6
- CMMLUKimi K2 Base by 4.486.590.9+4.4
- HellaSwagKimi K2 Base by 4.490.294.6+4.4
- MBPP+ (EvalPlus-augmented)GLM-4.5-Base by 4.37873.8+4.3
- MGSMKimi K2 Base by 3.981.385.2+3.9
- MMLU-ReduxKimi K2 Base by 3.686.690.2+3.6
- AGIEval-EnKimi K2 Base by 2.570.172.5+2.5
- BBHKimi K2 Base by 2.586.288.7+2.5
- GSM8KKimi K2 Base by 290.192.1+2
- CRUXEval-O (output prediction)Kimi K2 Base by 1.867.869.6+1.8
- MMMLUGLM-4.5-Base by 1.779.377.6+1.7
- CMathKimi K2 Base by 1.589.390.8+1.5
- OpenBookQAKimi K2 Base by 1.249.650.8+1.2
- DROPtie82.983.6tie
- Includetie76.375.7tie
- CRUXEval-I (input prediction)tie68.568tie
- Global-MMLU-Litetie85.885.6tie
- PIQAtie84.784.9tie
- RACEtie92.292tie
- ARC-Challengetie96.396.2tie
- MMLUtie87.787.8tie
- WinoGrandetie85.285.3tie
- HumanEvaltie78.278.2tie
Questions people ask
Which is better, GLM-4.5-Base or Kimi K2 Base?
GLM-4.5-Base and Kimi K2 Base each win one of the two areas where both have results. GLM-4.5-Base wins coding; Kimi K2 Base wins reasoning.
Which is better for coding?
GLM-4.5-Base. It wins the one coding test both models report.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 2 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 33 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.