Llama 3.3 Instruct 70B vs GPT-4o mini | grube.ai

Back to Benchmarks

Llama 3.3 Instruct 70B

Metavia Groq

vs

GPT-4o mini

OpenAI

Llama 3.3 Instruct 70B(Groq) wins 8 of 11

Index Scores

14.5

+1.9

Intelligence

12.6

10.7

+10.7

Coding

0.0

7.7

Math

+7.0

14.7

Benchmarks

71.3

+6.5

MMLU Pro

64.8

49.8

+7.2

GPQA

42.6

4.0

HLE

4.0

28.8

+5.4

LiveCode

23.4

26.0

+3.1

SciCode

22.9

77.3

MATH-500

+1.6

78.9

30.0

+18.3

AIME

11.7

Speed

144

+114

TPS

30

Shared

Winning delta

crafted by bart stefanski