Back to Benchmarks

Gemini 3.1 Pro Preview vs Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)

Gemini 3.1 Pro Preview

Googlevia Google AI Studio

vs

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)

Anthropic

Gemini 3.1 Pro Preview(Google AI Studio) wins 6 of 11

Index Scores

57.2
+5.5
Intelligence
51.7
55.5
+4.6
Coding
50.9
0.0
Math
0.0

Benchmarks

0.0
MMLU Pro
0.0
94.1
+6.6
GPQA
87.5
44.7
+14.7
HLE
30.0
0.0
LiveCode
0.0
58.9
+12.1
SciCode
46.8
0.0
MATH-500
0.0
0.0
AIME
0.0

Speed

71
+29
TPS
42
Shared
Winning delta

crafted by bart stefanski