GPT-5.3-Codex vs Qwen Plus 0728 (thinking)
Compare GPT-5.3-Codex by OpenAI against Qwen Plus 0728 (thinking) by Qwen, context windows of 400K vs 1.0M, tested across 37 shared challenges.
Our Verdict
GPT-5.3-Codex
Qwen Plus 0728 (thinking)Runner-up
No community votes yet. On paper, GPT-5.3-Codex has the edge — bigger model tier, newer, major provider backing.
Qwen Plus 0728 (thinking) is 3.5x cheaper per token — worth considering if cost matters.
Too close to call
API pricing
Cost per 1M tokens
GPT-5.3-Codex
Input
$1.75
Output
$14.00
Qwen Plus 0728 (thinking)
Input
$0.40
4.4× cheaper
Output
$4.00
3.5× cheaper
Qwen Plus 0728 (thinking) is cheaper on both — 4.4× input, 3.5× output
Writing DNA
Style Comparison
Similarity
66%
Qwen Plus 0728 (thinking) uses 3.2x more transitions
GPT-5.3-Codex
Qwen Plus 0728 (thinking)
65%Vocabulary54%
23wSentence Length15w
0.51Hedging0.36
6.5Bold7.0
6.2Lists4.6
0.16Emoji0.34
1.90Headings0.94
0.03Transitions0.09
Based on 23 + 18 text responses
Research
What we learned reading every model
FAQ
