DeepSeek-V3 vs Mistral Small 4
Wins 0 of 6 areas
—
Wins 4 of 6 areas
Coding · Reasoning · Long documents · Following instructions
Mistral Small 4 is the stronger all-rounder.
Scores updated · 21 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software02Mistral Small 42 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02Mistral Small 42 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01Mistral Small 41 of 1 test
- Following instructionsDoing exactly what it is asked01Mistral Small 41 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 3 ties
- FactsGetting facts right instead of making them up11Even1 each
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.
Mistral Small 4 costs 34% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Mistral Small 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+19.4points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+11.4points ahead
- Reasons across sets of long documentsAA-LCR+9points ahead
Where DeepSeek-V3 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3.8points ahead
Every test, side by side
All 21 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingMistral Small 4
- Terminal-Bench 2.1Mistral Small 4 by 4.116.921+4.1
- Terminal-Bench HardMistral Small 4 by 2.215.217.4+2.2
- SciCodetie3938.8tie
AgentsEven
- τ-Bench V3 · Bankingtie4.74.9tie
- GDPValtie00tie
- Terminal-Bench 4.0tie00tie
ReasoningMistral Small 4
- GPQA DiamondMistral Small 4 by 11.465.576.9+11.4
- Humanity's Last ExamMistral Small 4 by 5.24.79.9+5.2
- CritPttie00.3tie
FactsEven
- AA-Omniscience · Non-hallucinationMistral Small 4 by 19.414.133.5+19.4
- AA-Omniscience · AccuracyDeepSeek-V3 by 3.825.421.7+3.8
Following instructionsMistral Small 4
- IFBenchMistral Small 4 by 7.24148.2+7.2
Other results8 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.
- Arena HardDeepSeek-V3 by 33.191.458.3+33.1
- AIME 2025Mistral Small 4 by 32.551.383.8+32.5
- LiveCodeBenchMistral Small 4 by 14.449.263.6+14.4
- AA-OmniscienceMistral Small 4 by 10.3-40.7-30.4+10.3
- τ²-Bench Telecom (AA run)DeepSeek-V3 by 5.947.141.2+5.9
- Artificial Analysis Coding IndexMistral Small 4 by 3.62326.6+3.6
- MMLU-ProMistral Small 4 by 2.175.978+2.1
- AA IntelligenceMistral Small 4 by 1.69.711.3+1.6
Questions people ask
Which is better, DeepSeek-V3 or Mistral Small 4?
Mistral Small 4 wins four of the six areas where both have results: coding, reasoning, long documents and following instructions. DeepSeek-V3 wins none. They are level on agents and facts.
Which is better for coding?
Mistral Small 4. It wins 2 of the 3 coding tests both models report; DeepSeek-V3 wins none, and 1 is a tie.
Which is cheaper?
DeepSeek-V3 costs $0.24 per million input tokens and $0.90 per million output tokens; Mistral Small 4 costs $0.15 and $0.60. That makes Mistral Small 4 about 34% cheaper for the same work.
How do you compare the two?
We use the 21 benchmark tests both models have published scores on. The verdict counts the 13 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 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.