Llama3.3-70B-Instruct Benchmark Details

Llama3.3-70B-Instruct currently shows benchmark results led by MBPP (3 / 28, score 87.60), MATH (13 / 42, score 77), HumanEval (14 / 39, score 88.40).

Benchmark Results

Llama3.3-70B-Instruct

Benchmark Results

综合评估

3 evaluations

Benchmark / mode

Score

Rank/total

MMLU

Standard Mode

33 / 65

MMLU Pro

Standard Mode

68.90

92 / 124

GPQA Diamond

Standard Mode

50.50

148 / 175

编程与软件工程

3 evaluations

Benchmark / mode

Score

Rank/total

HumanEval

Standard Mode

88.40

14 / 39

MBPP

Standard Mode

87.60

3 / 28

LiveCodeBench

Standard Mode

33.30

107 / 118

数学推理

1 evaluations

Benchmark / mode

Score

Rank/total

MATH

Standard Mode

13 / 42

Compare with other models