Gemini 3 Pro vs MiniMax M2.5
Wins 3 of 6 areas
Coding · Reasoning · Long documents
Wins 3 of 6 areas
Agents · Facts · Following instructions
The two are evenly matched.Gemini 3 Pro is better at reasoning and coding; MiniMax M2.5 at facts and agents.
Scores updated · 27 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.
- ReasoningHard problems that need careful thinking40Gemini 3 Pro4 of 4 tests
- CodingWriting and fixing software32Gemini 3 Pro3 of 5 tests
- Long documentsFinding answers in very long texts10Gemini 3 Pro1 of 1 test
- FactsGetting facts right instead of making them up12MiniMax M2.52 of 3 tests
- AgentsCarrying out multi-step tasks on its own01MiniMax M2.51 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked01MiniMax M2.51 of 1 test
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.
MiniMax M2.5 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 Pro pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+29.6points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+26.2points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+19.2points ahead
Where MiniMax M2.5 pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+17.1points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+12.1points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+4points ahead
Every test, side by side
All 27 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3 Pro
- SciCodeGemini 3 Pro by 13.556.142.6+13.5
- SWE-bench ProMiniMax M2.5 by 12.143.355.4+12.1
- Terminal-Bench HardGemini 3 Pro by 6.941.734.8+6.9
- LMArena · WebDevGemini 3 Pro by 53 rating points14401387+53 rating
- SWE-bench VerifiedMiniMax M2.5 by 476.280.2+4
AgentsMiniMax M2.5
- BrowseCompMiniMax M2.5 by 17.159.276.3+17.1
- GDPValtie34.233.9tie
ReasoningGemini 3 Pro
- ARC-AGI-2Gemini 3 Pro by 26.231.14.9+26.2
- Humanity's Last ExamGemini 3 Pro by 19.239.720.5+19.2
- CritPtGemini 3 Pro by 89.11.1+8
- GPQA DiamondGemini 3 Pro by 690.884.8+6
FactsMiniMax M2.5
- AA-Omniscience · AccuracyGemini 3 Pro by 29.655.826.2+29.6
- Vectara HHEM hallucination ratelower is betterMiniMax M2.5 by 4.513.69.1+4.5
- AA-Omniscience · Non-hallucinationMiniMax M2.5 by 3.48.511.9+3.4
Following instructionsMiniMax M2.5
- IFBenchMiniMax M2.5 by 1.270.471.6+1.2
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.
- AA-OmniscienceGemini 3 Pro by 54.215.3-38.9+54.2
- ARC-AGI-1Gemini 3 Pro by 11.37563.7+11.3
- AIME25 no toolsGemini 3 Pro by 9.79686.3+9.7
- Artificial Analysis Coding IndexGemini 3 Pro by 9.146.537.4+9.1
- Multi-SWE-BenchMiniMax M2.5 by 8.642.751.3+8.6
- τ²-Bench Telecom (AA run)MiniMax M2.5 by 8.287.195.3+8.2
- AA IntelligenceGemini 3 Pro by 5.22822.8+5.2
- vectara_factual_consistencyMiniMax M2.5 by 4.586.490.9+4.5
- AA Agentic IndexMiniMax M2.5 by 3.65255.6+3.6
- vectara_avg_summary_lengthMiniMax M2.5 by 35.3 rating points101.9137.2+35.3 rating
- vectara_answer_rateGemini 3 Pro by 1.299.498.2+1.2
Questions people ask
Which is better, Gemini 3 Pro or MiniMax M2.5?
Gemini 3 Pro and MiniMax M2.5 each win three of the six areas where both have results. Gemini 3 Pro wins coding, reasoning and long documents; MiniMax M2.5 wins agents, facts and following instructions. MiniMax M2.5 costs 89% less.
Which is better for coding?
Gemini 3 Pro. It wins 3 of the 5 coding tests both models report; MiniMax M2.5 wins 2.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; MiniMax M2.5 costs $0.30 and $1.20. That makes MiniMax M2.5 about 89% cheaper for the same work.
How do you compare the two?
We use the 27 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 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.