Claude Sonnet 4.5 vs GPT-5.1-Codex-Mini
Compare Claude Sonnet 4.5 by Anthropic against GPT-5.1-Codex-Mini by OpenAI, in 7 community votes, gpt-5.1-codex-mini wins 57% of head-to-head duels, context windows of 200K vs 400K, tested across 43 shared challenges.
Reasoning: GPT-5.1-Codex-Mini wins 67% of votes
Image Generation: Claude Sonnet 4.5 wins 100% of votes
Our Verdict
GPT-5.1-Codex-MiniWinner
Claude Sonnet 4.5Runner-up
GPT-5.1-Codex-Mini has the edge overall. In 7 blind votes, GPT-5.1-Codex-Mini wins 57% of the time.
Pick Claude Sonnet 4.5 for Image Generation. Pick GPT-5.1-Codex-Mini for Reasoning.
Slight edge
API pricing
Cost per 1M tokens
Claude Sonnet 4.5
Input
$3.00
Output
$15.00
GPT-5.1-Codex-Mini
Input
$1.50
2.0× cheaper
Output
$6.00
2.5× cheaper
GPT-5.1-Codex-Mini is cheaper on both — 2.0× input, 2.5× output
Writing DNA
Style Comparison
Similarity
38%
Claude Sonnet 4.5 uses 45.2x more emoji
Claude Sonnet 4.5
GPT-5.1-Codex-Mini
64%Vocabulary65%
94wSentence Length20w
0.48Hedging0.56
4.9Bold3.3
8.0Lists3.4
0.45Emoji0.00
1.91Headings0.08
0.07Transitions0.05
Based on 20 + 13 text responses
Research
What we learned reading every model
FAQ
