Claude 3.7 Sonnet vs GPT-4
Compare Claude 3.7 Sonnet by Anthropic against GPT-4 by OpenAI, in 2 community votes, claude 3.7 sonnet wins 100% of head-to-head duels, context windows of 200K vs 8K, tested across 26 shared challenges.
Image Generation: Claude 3.7 Sonnet wins 100% of votes
Our Verdict
Claude 3.7 SonnetWinner
GPT-4Runner-up
Pick Claude 3.7 Sonnet. In 2 blind votes, Claude 3.7 Sonnet wins 100% of the time. That's not luck.
Claude 3.7 Sonnet particularly excels in Image Generation. Claude 3.7 Sonnet is 4.0x cheaper per token — worth considering if cost matters.
Clear winner
API pricing
Cost per 1M tokens
Claude 3.7 Sonnet
Input
$3.00
10× cheaper
Output
$15.00
4.0× cheaper
GPT-4
Input
$30.00
Output
$60.00
Claude 3.7 Sonnet is cheaper on both — 10× input, 4.0× output
Writing DNA
Style Comparison
Similarity
64%
Claude 3.7 Sonnet uses 177.5x more headings
Claude 3.7 Sonnet
GPT-4
62%Vocabulary59%
35wSentence Length18w
0.99Hedging1.10
1.2Bold0.8
4.3Lists2.5
0.00Emoji0.00
1.78Headings0.00
0.23Transitions0.45
Based on 13 + 15 text responses
Research
What we learned reading every model
FAQ
