Gemma 3 12B vs Ministral 3 8B
Wins 1 of 7 areas
Following instructions
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Long documents
Ministral 3 8B is the stronger all-rounder.Gemma 3 12B is cheaper and better at following instructions.
Scores updated · 26 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 software03Ministral 3 8B3 of 3 tests
- AgentsCarrying out multi-step tasks on its own01Ministral 3 8B1 of 3 tests · 2 ties
- ReasoningHard problems that need careful thinking01Ministral 3 8B1 of 3 tests · 2 ties
- FactsGetting facts right instead of making them up12Ministral 3 8B2 of 3 tests
- Images and chartsUnderstanding pictures, charts and video01Ministral 3 8B1 of 1 test
- Long documentsFinding answers in very long texts01Ministral 3 8B1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemma 3 12B1 of 1 test
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.
Gemma 3 12B costs 33% 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 Ministral 3 8B pulls ahead
- Reasons across sets of long documentsAA-LCR+17.4points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+12.2points ahead
- Harder college exam questions with imagesMMMU-Pro+8.5points ahead
Where Gemma 3 12B pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+7.6points ahead
Every test, side by side
All 26 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingMinistral 3 8B
- SciCodeMinistral 3 8B by 4.316.420.7+4.3
- Terminal-Bench 2.1Ministral 3 8B by 4.104.1+4.1
- Terminal-Bench HardMinistral 3 8B by 3.70.84.5+3.7
AgentsMinistral 3 8B
- τ-Bench V3 · BankingMinistral 3 8B by 2.90.83.7+2.9
- GDPValtie00tie
- Terminal-Bench 4.0tie00tie
ReasoningMinistral 3 8B
- GPQA DiamondMinistral 3 8B by 12.234.947.1+12.2
- Humanity's Last Examtie4.24.3tie
- CritPttie00tie
FactsMinistral 3 8B
- Vectara HHEM hallucination ratelower is betterGemma 3 12B by 17.34.421.7+17.3
- AA-Omniscience · Non-hallucinationMinistral 3 8B by 3.22.65.8+3.2
- AA-Omniscience · AccuracyMinistral 3 8B by 2.710.313+2.7
Images and chartsMinistral 3 8B
- MMMU-ProMinistral 3 8B by 8.537.546+8.5
Following instructionsGemma 3 12B
- IFBenchGemma 3 12B by 7.636.729.1+7.6
Other results11 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.
- LiveCodeBenchMinistral 3 8B by 3724.661.6+37
- vectara_factual_consistencyGemma 3 12B by 17.395.678.3+17.3
- τ²-Bench Telecom (AA run)Ministral 3 8B by 15.810.826.6+15.8
- AA-OmniscienceMinistral 3 8B by 8.2-77.1-69+8.2
- Arena HardMinistral 3 8B by 7.343.650.9+7.3
- vectara_avg_summary_lengthMinistral 3 8B by 49.7 rating points89.7139.4+49.7 rating
- Artificial Analysis Coding IndexMinistral 3 8B by 3.95.89.7+3.9
- WildBenchMinistral 3 8B by 3.663.266.8+3.6
- AA IntelligenceMinistral 3 8B by 1.73.85.5+1.7
- vectara_answer_rateMinistral 3 8B by 1.797.499.1+1.7
- MM MTBenchMinistral 3 8B by 1.46.78.1+1.4
Questions people ask
Which is better, Gemma 3 12B or Ministral 3 8B?
Ministral 3 8B wins six of the seven areas where both have results: coding, agents, reasoning, facts, images and charts and long documents. Gemma 3 12B wins following instructions, and costs 33% less.
Which is better for coding?
Ministral 3 8B. It wins 3 of the 3 coding tests both models report; Gemma 3 12B wins none.
Which is cheaper?
Gemma 3 12B costs $0.05 per million input tokens and $0.15 per million output tokens; Ministral 3 8B costs $0.15 and $0.15. That makes Gemma 3 12B about 33% cheaper for the same work.
How do you compare the two?
We use the 26 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 11 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.