Benchgen

AIME 2024 — Results

RankModelScore
1ripple-65b-preview97.1
2grok-3-mini0.958
3o4-mini0.934
4gemini-2-5-pro0.92
5o30.916
6deepseek-r1-05280.914
7glm-4-50.91
8glm-4-5-air0.894
9gemini-2-5-flash0.88
10o3-mini0.873
11qwen3-235b-a22b0.857
12qwen3-32b0.814
13phi-4-reasoning-plus0.813
14qwen3-30b-a3b0.804
15qwq-32b0.795
16kimi-k1-50.775
17phi-4-reasoning0.753
18o10.743
19kimi-k2-09050.72
20kimi-k2-instruct-09050.696
21kimi-k2-instruct0.696
22deepseek-v3-10.663
23deepseek-v3-03240.594
24gpt-4-1-mini0.496
25gpt-4-10.481

AIME 2024 — Leaderboard

1 phaseActive

AIME 2024 Leaderboard

No self-reported entries.