Benchgen

MMLU — Results

RankModelScore
1gpt-592.5
2o191.8
3gpt-4-590.8
4o1-preview90.8
5sarvam-105b90.6
6claude-3-5-sonnet-v190.4
7kimi-k2-090590.2
8kimi-k2-instruct89.5
9mimo-v2-5-pro89.4
10claude-3-5-sonnet88.7
11gpt-4o88.7
12llama-3-1-405b-instruct88.6
13deepseek-v388.5
14kimi-k2-base87.8
15qwen3-235b-a22b87.8
16grok-287.5
17kimi-k1-587.4
18o3-mini86.9
19claude-3-opus86.8
20qwen2-5-72b-instruct86.8
21gpt-4-turbo86.4
22gpt-4-061386.4
23grok-2-mini86.2
24llama-3-1-70b-instruct86
25llama-3-2-90b-instruct86
M

MMLU — Leaderboard

1 phaseActive

MMLU Leaderboard

No self-reported entries.