Llama3.1-70B-Instruct Benchmark Details

Llama3.1-70B-Instruct currently shows benchmark results led by MBPP (5 / 28, score 86), MMLU (33 / 66, score 86), HumanEval (23 / 39, score 80.50).

Benchmark Results

Llama3.1-70B-Instruct

Benchmark Results

General Knowledge

3 evaluations

Benchmark / mode

Score

Rank/total

86

33 / 66

66.40

103 / 132

48

164 / 187

Coding and Software Engineer

3 evaluations

Benchmark / mode

Score

Rank/total

86

5 / 28

80.50

23 / 39

33.30

112 / 123

Math and Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

67.80

26 / 42

Compare with other models