Mistral Devstral Medium
Nothing here. The model returned empty. We stared at it for a while.
Mistral Devstral Medium vs Mistral Devstral Small 1.1
Compare Mistral Devstral Medium and Mistral Devstral Small 1.1, both from Mistral AI, in 46 community votes, mistral devstral small 1.1 wins 61% of head-to-head duels, tested across 55 shared challenges. Updated April 2026.
Which is better, Mistral Devstral Medium or Mistral Devstral Small 1.1?
Mistral Devstral Small 1.1 is the better choice overall, winning 61% of 46 blind community votes on Rival. Mistral Devstral Medium costs $0.4/M input tokens vs $0.1/M for Mistral Devstral Small 1.1. Compare their real outputs side by side below.
Key Differences Between Mistral Devstral Medium and Mistral Devstral Small 1.1
Mistral Devstral Medium is made by mistral while Mistral Devstral Small 1.1 is from mistral. On pricing, Mistral Devstral Medium costs $0.4/M input tokens vs $0.1/M for Mistral Devstral Small 1.1. In community voting, In 46 community votes, Mistral Devstral Small 1.1 wins 61% of head-to-head duels.
In 46 community votes, Mistral Devstral Small 1.1 wins 61% of head-to-head duels. Mistral Devstral Medium leads in Reasoning, while Mistral Devstral Small 1.1 leads in Web Design. Based on blind community voting from the Rival open dataset of 46+ human preference judgments for this pair.
Web Design: Mistral Devstral Small 1.1 wins 72% of votes
Image Generation: Mistral Devstral Medium and Mistral Devstral Small 1.1 are tied
Reasoning: Mistral Devstral Medium wins 60% of votes
Conversation: Mistral Devstral Medium and Mistral Devstral Small 1.1 are tied
Our Verdict
Mistral Devstral Small 1.1Winner
Mistral Devstral MediumRunner-up
Pick Mistral Devstral Small 1.1. In 46 blind votes, Mistral Devstral Small 1.1 wins 61% of the time. That's not luck.
Pick Mistral Devstral Medium for Reasoning. Pick Mistral Devstral Small 1.1 for Web Design. Mistral Devstral Small 1.1 is 6.7x cheaper per token — worth considering if cost matters.
Clear winner
Writing DNA
Style Comparison
Similarity
98%
Mistral Devstral Medium uses 3.9x more headings
Mistral Devstral Medium
Mistral Devstral Small 1.1
50%Vocabulary50%
17wSentence Length17w
0.56Hedging0.56
4.5Bold4.3
5.3Lists5.7
0.00Emoji0.00
0.19Headings0.05
0.32Transitions0.34
Based on 21 + 21 text responses
vsAsk them anything yourself
279 AI models invented the same fake scientist.
We read every word. 250 models. 2.14 million words. This is what we found.
Free preview13 of 58 slides
FAQ