Grok 4 Fast (Reasoning) vs DeepSeek R1 Distill Llama 70B | grube.ai

Back to Benchmarks

Grok 4 Fast (Reasoning)

xAI

vs

DeepSeek R1 Distill Llama 70B

DeepSeekvia DeepInfra

Grok 4 Fast (Reasoning) wins 9 of 11

Index Scores

35.1

+19.1

Intelligence

16.0

27.4

+16.0

Coding

11.4

89.7

+36.0

Math

53.7

Benchmarks

85.0

+5.5

MMLU Pro

79.5

84.7

+44.5

GPQA

40.2

17.0

+10.9

HLE

6.1

83.2

+56.6

LiveCode

26.6

44.2

+13.0

SciCode

31.2

0.0

MATH-500

+93.5

93.5

0.0

AIME

+67.0

67.0

Speed

123

+81

TPS

42

Shared

Winning delta

crafted by bart stefanski