Atlas

Benchmarks

← All benchmarks

Kangaroo 2025 1-2

Math · 2025-10-20

This MathArena track evaluates models on the Kangaroo 2025 1-2 problem set. Scores report the percentage of problems answered correctly.

Top models (higher is better)

ModelScore
GPT-5.595.8
GPT-5.494.8
Kimi K393.8
Gemini 3.5 Flash89.6
Gemini 3 Flash Preview87.5
Gemini 3.1 Pro Preview86.5
Gemini 3.6 Flash83.3
Qwen3.5 397B A17B81.3
GPT-5.280.2
Gemini 3 Pro Preview76.0
Kimi K2.576.0
Opus 4.870.8
GPT-568.8
GLM-4.5V65.6
GPT-5.165.6
Gemini 2.5 Pro64.6
Sonnet 4.561.5
GPT-5 Mini61.5
Grok 461.5
Grok 4.1 Fast60.4
Opus 4.659.4
Grok 4 Fast58.3
Qwen3-VL-235B-A22B-Instruct58.3
Loading Atlas data…