Devstral Small 2 vs Mistral Small 3.2
Wins 4 of 7 areas
Coding · Agents · Reasoning · Long documents
Wins 2 of 7 areas
Images and charts · Following instructions
Devstral Small 2 wins more areas, narrowly.Mistral Small 3.2 is better at images and charts.
Scores updated · 18 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software30Devstral Small 23 of 3 tests
- AgentsCarrying out multi-step tasks on its own20Devstral Small 22 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking10Devstral Small 21 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts10Devstral Small 21 of 1 test
- Images and chartsUnderstanding pictures, charts and video01Mistral Small 3.21 of 1 test
- Following instructionsDoing exactly what it is asked01Mistral Small 3.21 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Images and charts, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Devstral Small 2 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+24points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+9.9points ahead
- Reasons across sets of long documentsAA-LCR+7.7points ahead
Where Mistral Small 3.2 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+10.5points ahead
- Harder college exam questions with imagesMMMU-Pro+3.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+2.3points ahead
Every test, side by side
All 18 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDevstral Small 2
- Terminal-Bench 2.1Devstral Small 2 by 2429.65.6+24
- Terminal-Bench HardDevstral Small 2 by 9.916.76.8+9.9
- SciCodeDevstral Small 2 by 3.832.428.6+3.8
AgentsDevstral Small 2
- τ-Bench V3 · BankingDevstral Small 2 by 4.510.76.2+4.5
- GDPValDevstral Small 2 by 1.61.60+1.6
- Terminal-Bench 4.0tie00tie
ReasoningDevstral Small 2
- GPQA DiamondDevstral Small 2 by 2.753.250.5+2.7
- Humanity's Last Examtie3.54.3tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationMistral Small 3.2 by 10.512.723.2+10.5
- AA-Omniscience · AccuracyDevstral Small 2 by 115.814.8+1
Images and chartsMistral Small 3.2
- MMMU-ProMistral Small 3.2 by 3.444.648+3.4
Following instructionsMistral Small 3.2
- IFBenchMistral Small 3.2 by 2.331.233.5+2.3
Other results4 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- Artificial Analysis Coding IndexDevstral Small 2 by 16.829.312.5+16.8
- AA-OmniscienceMistral Small 3.2 by 7.1-57.7-50.6+7.1
- τ²-Bench Telecom (AA run)Mistral Small 3.2 by 6.123.429.5+6.1
- AA Intelligencetie7.58.2tie
Questions people ask
Which is better, Devstral Small 2 or Mistral Small 3.2?
Devstral Small 2 wins four of the seven areas where both have results: coding, agents, reasoning and long documents. Mistral Small 3.2 wins images and charts and following instructions. They are level on facts.
Which is better for coding?
Devstral Small 2. It wins 3 of the 3 coding tests both models report; Mistral Small 3.2 wins none.
How do you compare the two?
We use the 18 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 4 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.