GPT-4o mini vs Qwen Plus 0728 (thinking)
Compare GPT-4o mini by OpenAI against Qwen Plus 0728 (thinking) by Qwen, context windows of 128K vs 1.0M, tested across 37 shared challenges.
Our Verdict
Qwen Plus 0728 (thinking)
GPT-4o miniRunner-up
No community votes yet. On paper, Qwen Plus 0728 (thinking) has the edge — bigger model tier, newer, bigger context window.
GPT-4o mini is 6.7x cheaper per token — worth considering if cost matters.
Too close to call
API pricing
Cost per 1M tokens
GPT-4o mini
Input
$0.15
2.7× cheaper
Output
$0.60
6.7× cheaper
Qwen Plus 0728 (thinking)
Input
$0.40
Output
$4.00
GPT-4o mini is cheaper on both — 2.7× input, 6.7× output
Writing DNA
Style Comparison
Similarity
31%
Qwen Plus 0728 (thinking) uses 34.0x more emoji
GPT-4o mini
Qwen Plus 0728 (thinking)
54%Vocabulary54%
18wSentence Length15w
0.84Hedging0.36
3.8Bold7.0
3.3Lists4.6
0.00Emoji0.34
1.03Headings0.94
0.23Transitions0.09
Based on 22 + 18 text responses
Research
What we learned reading every model
FAQ
