DataLearner logo

Claude 3.5 Haiku Benchmark Details

Claude 3.5 Haiku currently shows benchmark results led by MBPP (6 / 28, score 85.60), HumanEval (17 / 39, score 88.10), MATH (23 / 42, score 69.20).

Benchmark Results

Claude 3.5 Haiku

Benchmark Results

Thinking

General Knowledge

4 evaluations
Benchmark / mode
Score
Rank/total
MMLU
77.60
54 / 66
MMLU Pro
65
105 / 132
GPQA Diamond
41.60
171 / 188
GPQA
37.50
13 / 15

Coding and Software Engineer

2 evaluations
Benchmark / mode
Score
Rank/total
HumanEval
88.10
17 / 39
MBPP
85.60
6 / 28

Math and Reasoning

2 evaluations
Benchmark / mode
Score
Rank/total
MATH
69.20
23 / 42
FrontierMath
0.30
57 / 60

Common Sense

1 evaluations
Benchmark / mode
Score
Rank/total
SimpleQA
8.02
44 / 47

Agent Level Benchmark

1 evaluations
Benchmark / mode
Score
Rank/total
Aider-Polyglot
Standard Mode
28
45 / 59
Compare with other models