Gemini 3.1 Pro vs MiniMax M3
Wins 5 of 8 areas
Coding · Agents · Reasoning · Images and charts · Math
Wins 1 of 8 areas
Long documents
Gemini 3.1 Pro is the stronger all-rounder.MiniMax M3 is cheaper and better at long documents.
Scores updated · 47 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking50Gemini 3.1 Pro5 of 5 tests
- CodingWriting and fixing software52Gemini 3.1 Pro5 of 8 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own52Gemini 3.1 Pro5 of 7 tests
- Images and chartsUnderstanding pictures, charts and video30Gemini 3.1 Pro3 of 3 tests
- MathCompetition and research-level math20Gemini 3.1 Pro2 of 2 tests
- Long documentsFinding answers in very long texts01MiniMax M31 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
- Following instructionsDoing exactly what it is asked11Even1 each
Long documents rests on a single test.
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.
MiniMax M3 costs 89% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 3.1 Pro pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+38.2points ahead
- Common-sense trick questionsSimpleBench+33.8points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+21.6points ahead
Where MiniMax M3 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+32.5points ahead
- Real work tasks from 44 professionsGDPVal+22.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+5.8points ahead
Every test, side by side
All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3.1 Pro
- SciCodeGemini 3.1 Pro by 11.658.747.1+11.6
- Terminal-Bench HardGemini 3.1 Pro by 11.453.842.4+11.4
- Terminal-Bench 2.1Gemini 3.1 Pro by 8.673.865.2+8.6
- LiveBench · CodingGemini 3.1 Pro by 8.376.568.2+8.3
- SWE-bench ProMiniMax M3 by 4.854.259+4.8
- LMArena · WebDevMiniMax M3 by 35 rating points14471482+35 rating
- LiveBench · Agentic CodingGemini 3.1 Pro by 3.444.140.7+3.4
- SWE-bench Verifiedtie80.680.5tie
AgentsGemini 3.1 Pro
- GDPValMiniMax M3 by 22.414.737.1+22.4
- OSWorld-VerifiedGemini 3.1 Pro by 6.176.270.1+6.1
- τ-Bench V3 · BankingGemini 3.1 Pro by 6.121.415.3+6.1
- MCP AtlasMiniMax M3 by 569.274.2+5
- AA ApexAgentsGemini 3.1 Pro by 4.33227.7+4.3
- BrowseCompGemini 3.1 Pro by 2.485.983.5+2.4
- Terminal-Bench 4.0Gemini 3.1 Pro by 242+2
ReasoningGemini 3.1 Pro
- SimpleBenchGemini 3.1 Pro by 33.879.645.8+33.8
- CritPtGemini 3.1 Pro by 1417.73.7+14
- LiveBench · ReasoningGemini 3.1 Pro by 9.58474.5+9.5
- Humanity's Last ExamGemini 3.1 Pro by 84739+8
- GPQA DiamondGemini 3.1 Pro by 1.294.192.9+1.2
FactsEven
- AA-Omniscience · AccuracyGemini 3.1 Pro by 38.254.916.7+38.2
- AA-Omniscience · Non-hallucinationMiniMax M3 by 32.549.181.6+32.5
Images and chartsGemini 3.1 Pro
- LMArena · VisionGemini 3.1 Pro by 42 rating points12961254+42 rating
- MMMU-ProGemini 3.1 Pro by 3.882.478.6+3.8
- Video-MMEGemini 3.1 Pro by 1.386.785.4+1.3
MathGemini 3.1 Pro
- LiveBench · MathematicsGemini 3.1 Pro by 149177+14
- HMMT Feb. 2026Gemini 3.1 Pro by 10.394.784.4+10.3
Following instructionsEven
- LiveBench · Instruction FollowingGemini 3.1 Pro by 21.679.157.5+21.6
- IFBenchMiniMax M3 by 5.877.182.9+5.8
Other results17 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.
- GDPval-AA v2MiniMax M3 by 429 rating points9621391+429 rating
- AA-OmniscienceGemini 3.1 Pro by 30.531.91.4+30.5
- OfficeQA ProGemini 3.1 Pro by 27.472.545.1+27.4
- AA Agentic IndexMiniMax M3 by 20.510.330.8+20.5
- PostTrainBenchMiniMax M3 by 15.521.637.1+15.5
- HMMT Nov. 2025Gemini 3.1 Pro by 10.494.884.4+10.4
- Artificial Analysis Coding IndexGemini 3.1 Pro by 10.268.858.6+10.2
- DeepSWEMiniMax M3 by 101020+10
- LiveBenchGemini 3.1 Pro by 9.979.970+9.9
- NL2RepoMiniMax M3 by 8.733.442.1+8.7
- livebench_languageGemini 3.1 Pro by 8.685.476.8+8.6
- τ²-Bench Telecom (AA run)Gemini 3.1 Pro by 6.795.688.9+6.7
- Finance Agent v2MiniMax M3 by 5.34348.3+5.3
- Tool-DecathlonMiniMax M3 by 448.852.8+4
- τ³-Bench BankingGemini 3.1 Pro by 3.516.513+3.5
- livebench_data_analysisGemini 3.1 Pro by 2.378.576.2+2.3
- AA Intelligencetie29.729.2tie
Questions people ask
Which is better, Gemini 3.1 Pro or MiniMax M3?
Gemini 3.1 Pro wins five of the eight areas we test: coding, agents, reasoning, images and charts and math. MiniMax M3 wins long documents, and costs 89% less. They are level on facts and following instructions.
Which is better for coding?
Gemini 3.1 Pro. It wins 5 of the 8 coding tests both models report; MiniMax M3 wins 2, and 1 is a tie.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; MiniMax M3 costs $0.30 and $1.20. That makes MiniMax M3 about 89% cheaper for the same work.
How do you compare the two?
We use the 47 benchmark tests both models have published scores on. The verdict counts the 30 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.