Claude 3.7 Thinking Sonnet vs DeepSeek V3 (March 2024)
Compare Claude 3.7 Thinking Sonnet by Anthropic against DeepSeek V3 (March 2024) by DeepSeek, in 15 community votes, claude 3.7 thinking sonnet wins 62% of head-to-head duels, context windows of 200K vs 128K, tested across 53 shared challenges.
Web Design: Claude 3.7 Thinking Sonnet and DeepSeek V3 (March 2024) are tied
Reasoning: Claude 3.7 Thinking Sonnet and DeepSeek V3 (March 2024) are tied
Image Generation: Claude 3.7 Thinking Sonnet wins 100% of votes
Our Verdict
Claude 3.7 Thinking SonnetWinner
DeepSeek V3 (March 2024)Runner-up
Pick Claude 3.7 Thinking Sonnet. In 15 blind votes, Claude 3.7 Thinking Sonnet wins 62% of the time. That's not luck.
Claude 3.7 Thinking Sonnet particularly excels in Image Generation. DeepSeek V3 (March 2024) is 107x cheaper per token — worth considering if cost matters.
Clear winner
API pricing
Cost per 1M tokens
Claude 3.7 Thinking Sonnet
Input
$6.00
Output
$30.00
DeepSeek V3 (March 2024)
Input
$0.14
43× cheaper
Output
$0.28
107× cheaper
DeepSeek V3 (March 2024) is cheaper on both — 43× input, 107× output
Writing DNA
Style Comparison
Similarity
54%
DeepSeek V3 (March 2024) uses 25.6x more emoji
Claude 3.7 Thinking Sonnet
DeepSeek V3 (March 2024)
61%Vocabulary58%
49wSentence Length18w
0.64Hedging0.73
2.2Bold7.5
4.5Lists7.6
0.00Emoji0.26
1.81Headings1.60
0.21Transitions0.08
Based on 21 + 16 text responses
Research
What we learned reading every model
FAQ
