Devstral Small 2 vs GPT-5 mini
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions
GPT-5 mini is the stronger all-rounder.
Scores updated · 20 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.
- AgentsCarrying out multi-step tasks on its own02GPT-5 mini2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02GPT-5 mini2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up02GPT-5 mini2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video01GPT-5 mini1 of 1 test
- Long documentsFinding answers in very long texts01GPT-5 mini1 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5 mini1 of 1 test
- CodingWriting and fixing software22Even2 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 tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where GPT-5 mini pulls ahead
- Reasons across sets of long documentsAA-LCR+44.3points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+44.2points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+30.9points ahead
Where Devstral Small 2 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+25.9points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+8.2points ahead
Every test, side by side
All 20 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- Terminal-Bench 2.1Devstral Small 2 by 25.929.63.7+25.9
- Terminal-Bench HardGPT-5 mini by 16.616.733.3+16.6
- SWE-bench VerifiedDevstral Small 2 by 8.26859.8+8.2
- SciCodeGPT-5 mini by 6.632.439+6.6
AgentsGPT-5 mini
- GDPValGPT-5 mini by 11.91.613.5+11.9
- τ-Bench V3 · BankingGPT-5 mini by 4.810.715.5+4.8
- Terminal-Bench 4.0tie00tie
ReasoningGPT-5 mini
- GPQA DiamondGPT-5 mini by 29.653.282.8+29.6
- Humanity's Last ExamGPT-5 mini by 183.521.5+18
- CritPttie00tie
FactsGPT-5 mini
- AA-Omniscience · Non-hallucinationGPT-5 mini by 30.912.743.6+30.9
- AA-Omniscience · AccuracyGPT-5 mini by 9.215.825+9.2
Following instructionsGPT-5 mini
- IFBenchGPT-5 mini by 44.231.275.4+44.2
Other results5 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.
- τ²-Bench Telecom (AA run)GPT-5 mini by 4523.468.4+45
- AA-OmniscienceGPT-5 mini by 40.4-57.7-17.3+40.4
- Artificial Analysis Coding IndexDevstral Small 2 by 13.729.315.6+13.7
- AA IntelligenceGPT-5 mini by 9.37.516.8+9.3
- AA Agentic IndexGPT-5 mini by 4.14.88.9+4.1
Questions people ask
Which is better, Devstral Small 2 or GPT-5 mini?
GPT-5 mini wins six of the seven areas where both have results: agents, reasoning, facts, images and charts, long documents and following instructions. Devstral Small 2 wins none. They are level on coding.
Which is better for coding?
Neither. They win 2 coding tests each of the 4 both models report.
How do you compare the two?
We use the 20 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 5 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.