Claude 3.7 Thinking Sonnet vs OpenAI o3
Compare Claude 3.7 Thinking Sonnet by Anthropic against OpenAI o3 by OpenAI, in 41 community votes, claude 3.7 thinking sonnet wins 79% of head-to-head duels, tested across 54 shared challenges.
Web Design: Claude 3.7 Thinking Sonnet wins 73% of votes
Image Generation: Claude 3.7 Thinking Sonnet wins 83% of votes
Reasoning: Claude 3.7 Thinking Sonnet wins 80% of votes
Conversation: Claude 3.7 Thinking Sonnet wins 100% of votes
Our Verdict
Claude 3.7 Thinking SonnetWinner
OpenAI o3Runner-up
Pick Claude 3.7 Thinking Sonnet. In 41 blind votes, Claude 3.7 Thinking Sonnet wins 79% of the time. That's not luck.
Claude 3.7 Thinking Sonnet particularly excels in Conversation, Image Generation, Reasoning.
Clear winner
API pricing
Cost per 1M tokens
Claude 3.7 Thinking Sonnet
Input
$6.00
1.7× cheaper
Output
$30.00
1.3× cheaper
OpenAI o3
Input
$10.00
Output
$40.00
Claude 3.7 Thinking Sonnet is cheaper on both — 1.7× input, 1.3× output
Writing DNA
Style Comparison
Similarity
45%
Claude 3.7 Thinking Sonnet uses 4.8x more headings
Claude 3.7 Thinking Sonnet
OpenAI o3
61%Vocabulary68%
49wSentence Length14w
0.64Hedging0.26
2.2Bold0.7
4.5Lists3.0
0.00Emoji0.00
1.81Headings0.38
0.21Transitions0.12
Based on 21 + 16 text responses
Research
What we learned reading every model
FAQ
