GPT-5.4 mini (xhigh) vs DeepSeek R1 Distill Llama 70B | grube.ai

Back to Benchmarks

GPT-5.4 mini (xhigh)

OpenAI

vs

DeepSeek R1 Distill Llama 70B

DeepSeekvia DeepInfra

GPT-5.4 mini (xhigh) wins 6 of 11

Index Scores

48.1

+32.1

Intelligence

16.0

51.5

+40.1

Coding

11.4

0.0

Math

+53.7

53.7

Benchmarks

0.0

MMLU Pro

+79.5

79.5

87.5

+47.3

GPQA

40.2

26.6

+20.5

HLE

6.1

0.0

LiveCode

+26.6

26.6

49.9

+18.7

SciCode

31.2

0.0

MATH-500

+93.5

93.5

0.0

AIME

+67.0

67.0

Speed

83

+41

TPS

42

Shared

Winning delta

crafted by bart stefanski