Command A+ vs Gemma 4 31B
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Gemma 4 31B 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 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 software03Gemma 4 31B3 of 3 tests
- ReasoningHard problems that need careful thinking03Gemma 4 31B3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video13Gemma 4 31B3 of 4 tests
- AgentsCarrying out multi-step tasks on its own01Gemma 4 31B1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Gemma 4 31B1 of 1 test
- Following instructionsDoing exactly what it is asked01Gemma 4 31B1 of 1 test
- 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.
Gemma 4 31B costs 69% 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 Gemma 4 31B pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+20.6points ahead
- Reasons across sets of long documentsAA-LCR+17points ahead
- Reasoning about charts from research papersCharXiv (RQ)+15.2points ahead
Where Command A+ pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+70.8points ahead
- Math problems shown in pictures and chartsMathVista+1.3points 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.
CodingGemma 4 31B
- Terminal-Bench 2.1Gemma 4 31B by 20.622.843.4+20.6
- Terminal-Bench HardGemma 4 31B by 11.42536.4+11.4
- SciCodeGemma 4 31B by 738.545.5+7
AgentsGemma 4 31B
- τ-Bench V3 · BankingGemma 4 31B by 8.8614.8+8.8
- Terminal-Bench 4.0tie0.50tie
ReasoningGemma 4 31B
- Humanity's Last ExamGemma 4 31B by 11.61223.6+11.6
- GPQA DiamondGemma 4 31B by 9.676.185.7+9.6
- CritPtGemma 4 31B by 1.10.31.4+1.1
FactsEven
- AA-Omniscience · Non-hallucinationCommand A+ by 70.885.815+70.8
- AA-Omniscience · AccuracyGemma 4 31B by 11.18.920+11.1
Images and chartsGemma 4 31B
- CharXiv (RQ)Gemma 4 31B by 15.252.767.9+15.2
- MMMU-ProGemma 4 31B by 10.263.273.4+10.2
- MMMUGemma 4 31B by 5.375.180.4+5.3
- MathVistaCommand A+ by 1.380.679.3+1.3
Following instructionsGemma 4 31B
- IFBenchGemma 4 31B by 1.773.975.6+1.7
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.
- AA-OmniscienceCommand A+ by 43.9-4-47.9+43.9
- τ²-Bench Telecom (AA run)Command A+ by 20.880.759.9+20.8
- Artificial Analysis Coding IndexGemma 4 31B by 15.627.843.4+15.6
- AA Agentic IndexGemma 4 31B by 3.13.66.7+3.1
- AA IntelligenceGemma 4 31B by 1.613.114.7+1.6
Questions people ask
Which is better, Command A+ or Gemma 4 31B?
Gemma 4 31B wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Command A+ wins none. They are level on facts.
Which is better for coding?
Gemma 4 31B. It wins 3 of the 3 coding tests both models report; Command A+ wins none.
Which is cheaper?
Command A+ costs $0.30 per million input tokens and $1.50 per million output tokens; Gemma 4 31B costs $0.17 and $0.40. That makes Gemma 4 31B about 69% 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 16 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.