GPT-4 vs Qwen Plus 0728 (thinking)
Compare GPT-4 by OpenAI against Qwen Plus 0728 (thinking) by Qwen, context windows of 8K vs 1.0M, tested across 23 shared challenges.
Our Verdict
GPT-4
Qwen Plus 0728 (thinking)
No community votes yet. On paper, these are closely matched - try both with your actual task to see which fits your workflow.
Qwen Plus 0728 (thinking) is 15x cheaper per token — worth considering if cost matters.
Too close to call
API pricing
Cost per 1M tokens
GPT-4
Input
$30.00
Output
$60.00
Qwen Plus 0728 (thinking)
Input
$0.40
75× cheaper
Output
$4.00
15× cheaper
Qwen Plus 0728 (thinking) is cheaper on both — 75× input, 15× output
Writing DNA
Style Comparison
Similarity
38%
Qwen Plus 0728 (thinking) uses 34.0x more emoji
GPT-4
Qwen Plus 0728 (thinking)
59%Vocabulary54%
18wSentence Length15w
1.10Hedging0.36
0.8Bold7.0
2.5Lists4.6
0.00Emoji0.34
0.00Headings0.94
0.45Transitions0.09
Based on 15 + 18 text responses
Research
What we learned reading every model
FAQ
