Back to Benchmarks

Claude Opus 4.7 (Adaptive Reasoning, Max Effort) vs gpt-oss-120B (high)

Claude Opus 4.7 (Adaptive Reasoning, Max Effort)

Anthropic

vs

gpt-oss-120B (high)

OpenAIvia Groq

Claude Opus 4.7 (Adaptive Reasoning, Max Effort) wins 5 of 11

Index Scores

57.3
+24.0
Intelligence
33.3
52.5
+23.9
Coding
28.6
0.0
Math
+93.4
93.4

Benchmarks

0.0
MMLU Pro
+80.8
80.8
91.4
+13.2
GPQA
78.2
39.6
+21.1
HLE
18.5
0.0
LiveCode
+87.8
87.8
54.5
+15.6
SciCode
38.9
0.0
MATH-500
0.0
0.0
AIME
0.0

Speed

52
TPS
+297
349
Shared
Winning delta

crafted by bart stefanski