GPT OSS 20B vs Mistral Large 2
Compare GPT OSS 20B by OpenAI against Mistral Large 2 by Mistral AI, context windows of 131K vs 128K, tested across 23 shared challenges.
Our Verdict
GPT OSS 20B
Mistral Large 2Runner-up
No community votes yet. On paper, GPT OSS 20B has the edge — newer, major provider backing.
GPT OSS 20B is 240x cheaper per token — worth considering if cost matters.
Too close to call
API pricing
Cost per 1M tokens
GPT OSS 20B
Input
$0.02
400× cheaper
Output
$0.10
240× cheaper
Mistral Large 2
Input
$8.00
Output
$24.00
GPT OSS 20B is cheaper on both — 400× input, 240× output
Writing DNA
Style Comparison
Similarity
62%
GPT OSS 20B uses 31.7x more transitions
GPT OSS 20B
Mistral Large 2
54%Vocabulary43%
17wSentence Length21w
0.26Hedging0.39
5.9Bold14.4
3.3Lists7.4
0.10Emoji0.80
0.75Headings1.41
0.32Transitions0.01
Based on 21 + 10 text responses
Research
What we learned reading every model
FAQ
