Back to Benchmarks

Gemma 4 31B (Reasoning) vs GPT-4.1

Gemma 4 31B (Reasoning)

Googlevia DeepInfra

vs

GPT-4.1

OpenAI

GPT-4.1 wins 6 of 11

Index Scores

39.2
+12.9
Intelligence
26.3
38.7
+16.9
Coding
21.8
0.0
Math
+34.7
34.7

Benchmarks

0.0
MMLU Pro
+80.6
80.6
85.7
+19.1
GPQA
66.6
22.7
+18.1
HLE
4.6
0.0
LiveCode
+45.7
45.7
43.4
+5.3
SciCode
38.1
0.0
MATH-500
+91.3
91.3
0.0
AIME
+43.7
43.7

Speed

32
TPS
+12
44
Shared
Winning delta

crafted by bart stefanski