Model Comparison

O3-mini vs Claude Sonnet 3.7

"Claude 3.7 Sonnet is roughly 3.3x more expensive than o3-mini for tokens."

o3-mini Input: $1.10/million tokens
Claude Input: $3.00/million tokens
o3-mini Output: $4.40/million tokens
Claude Output: $15.00/million tokens

"Claude 3.7 Sonnet excels in reasoning with 84.8% accuracy in GPQA."

Claude GPQA: 84.8%
o3-mini GPQA: Not specified

"Claude 3.7 Sonnet achieves 96.2% in MATH benchmarks."

Claude MATH: 96.2%
Claude Context: 200K tokens
o3-mini Context: 200K tokens

"O3-Mini-High outperforms Claude 3.7 Sonnet in identifying critical code issues."

o3-mini Output: 100K tokens
Claude Output: 128K tokens