DataLearner logo

Llama 4 Maverick Benchmark Details

Llama 4 Maverick currently shows benchmark results led by MBPP (13 / 28, score 77.60), MMLU (38 / 66, score 85.50), MATH (30 / 42, score 61.20). 1 source link is attached for reference.

Benchmark Results

Llama 4 Maverick

Benchmark Results

Thinking
Tool usage

General Knowledge

4 evaluations
Benchmark / mode
Score
Rank/total
MMLU
85.50
38 / 66
MMLU Pro
62.90
107 / 132
ARC-AGI
Thinking Enabled
4.40
66 / 68
ARC-AGI-2
Thinking Enabled
0
60 / 62

Coding and Software Engineer

1 evaluations
Benchmark / mode
Score
Rank/total
MBPP
77.60
13 / 28

Math and Reasoning

2 evaluations
Benchmark / mode
Score
Rank/total
MATH
61.20
30 / 42
FrontierMath
0.70
55 / 60

Common Sense Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
Simple Bench
Standard Mode
27.70
49 / 63

Agent Level Benchmark

1 evaluations
Benchmark / mode
Score
Rank/total
Aider-Polyglot
Standard Mode
15.60
54 / 59

Claw-style Agent Evaluation

1 evaluations
Benchmark / mode
Score
Rank/total
Pinch Bench
Thinking EnabledTools
46.10
36 / 37
Compare with other models

Sources

arcprize.orgarcprize.org