Llama 4 Behemoth Instruct Benchmark Details

Llama 4 Behemoth Instruct currently shows benchmark results led by MMLU Pro (49 / 126, score 82.20), GPQA Diamond (98 / 179, score 73.70), MATH-500 (25 / 44, score 95).

Benchmark Results

Llama 4 Behemoth Instruct

Benchmark Results

Thinking

General Knowledge

2 evaluations
Benchmark / mode
Score
Rank/total
82.20
49 / 126
73.70
98 / 179

Math and Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
95
25 / 44

Coding and Software Engineer

1 evaluations
Benchmark / mode
Score
Rank/total
49.40
92 / 120