Claude Sonnet 3.6 (2022-10-22) vs GPT-5 Codex

Compare Claude Sonnet 3.6 (2022-10-22) by Anthropic against GPT-5 Codex by OpenAI, tested across 50 shared challenges. Updated April 2026.

Which is better, Claude Sonnet 3.6 (2022-10-22) or GPT-5 Codex?

Claude Sonnet 3.6 (2022-10-22) and GPT-5 Codex are both competitive models. Compare their real outputs side by side below.

Key Differences Between Claude Sonnet 3.6 (2022-10-22) and GPT-5 Codex

Claude Sonnet 3.6 (2022-10-22) is made by anthropic while GPT-5 Codex is from openai.

Our Verdict
GPT-5 Codex
GPT-5 Codex
Claude Sonnet 3.6 (2022-10-22)
Claude Sonnet 3.6 (2022-10-22)Runner-up

No community votes yet. On paper, GPT-5 Codex has the edge — bigger model tier, newer.

Too close to call
Writing DNA

Style Comparison

Similarity
93%

GPT-5 Codex uses 14.6x more emoji

Claude Sonnet 3.6 (2022-10-22)
GPT-5 Codex
66%Vocabulary64%
74wSentence Length17w
0.72Hedging0.38
0.3Bold2.3
8.9Lists3.5
0.00Emoji0.15
0.06Headings0.49
0.09Transitions0.05
Based on 21 + 13 text responses
vs

Ask them anything yourself

Claude Sonnet 3.6 (2022-10-22)GPT-5 Codex

279 AI models invented the same fake scientist.

We read every word. 250 models. 2.14 million words. This is what we found.

AI Hallucination Index 2026
Free preview13 of 58 slides
FAQ

Common questions