Gemini 3.1 Pro vs GPT-5.4 mini
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
Gemini 3.1 Pro is the stronger all-rounder.GPT-5.4 mini is cheaper.
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.
- CodingWriting and fixing software60Gemini 3.1 Pro6 of 7 tests · 1 tie
- ReasoningHard problems that need careful thinking50Gemini 3.1 Pro5 of 5 tests
- MathCompetition and research-level math30Gemini 3.1 Pro3 of 3 tests
- FactsGetting facts right instead of making them up31Gemini 3.1 Pro3 of 4 tests
- Images and chartsUnderstanding pictures, charts and video20Gemini 3.1 Pro2 of 2 tests
- Following instructionsDoing exactly what it is asked20Gemini 3.1 Pro2 of 2 tests
- AgentsCarrying out multi-step tasks on its own43Gemini 3.1 Pro4 of 7 tests
- Long documentsFinding answers in very long texts10Gemini 3.1 Pro1 of 1 test
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.
GPT-5.4 mini costs 63% 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
- Abstract visual puzzles that people can solveARC-AGI-2+58.2points ahead
- Short factual questions, answered correctlySimpleQA Verified+44.1points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+39.3points ahead
Where GPT-5.4 mini pulls ahead
- Real work tasks from 44 professionsGDPVal+11.1points ahead
- Finds the root cause of IT system incidentsAA IT-Bench SRE+4.9points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+4.2points 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
- Terminal-Bench 2.1Gemini 3.1 Pro by 14.673.859.2+14.6
- SciCodeGemini 3.1 Pro by 6.658.752.1+6.6
- LMArena · WebDevGemini 3.1 Pro by 50 rating points14471397+50 rating
- LiveBench · CodingGemini 3.1 Pro by 4.976.571.6+4.9
- LiveBench · Agentic CodingGemini 3.1 Pro by 2.444.141.7+2.4
- Terminal-Bench HardGemini 3.1 Pro by 1.553.852.3+1.5
- SWE-bench Protie54.254.4tie
AgentsGemini 3.1 Pro
- MCP AtlasGemini 3.1 Pro by 11.569.257.7+11.5
- GDPValGPT-5.4 mini by 11.114.725.8+11.1
- AA IT-Bench SREGPT-5.4 mini by 4.930.335.2+4.9
- τ-Bench V3 · BankingGPT-5.4 mini by 4.221.425.6+4.2
- OSWorld-VerifiedGemini 3.1 Pro by 4.176.272.1+4.1
- AA ApexAgentsGemini 3.1 Pro by 3.83228.2+3.8
- Terminal-Bench 4.0Gemini 3.1 Pro by 242+2
ReasoningGemini 3.1 Pro
- ARC-AGI-2Gemini 3.1 Pro by 58.277.118.9+58.2
- Humanity's Last ExamGemini 3.1 Pro by 18.94728.1+18.9
- LiveBench · ReasoningGemini 3.1 Pro by 12.78471.3+12.7
- CritPtGemini 3.1 Pro by 7.717.710+7.7
- GPQA DiamondGemini 3.1 Pro by 6.694.187.5+6.6
FactsGemini 3.1 Pro
- SimpleQA VerifiedGemini 3.1 Pro by 44.173.529.4+44.1
- AA-Omniscience · Non-hallucinationGemini 3.1 Pro by 39.349.19.8+39.3
- AA-Omniscience · AccuracyGemini 3.1 Pro by 17.454.937.5+17.4
- Vectara HHEM hallucination ratelower is betterGPT-5.4 mini by 4.910.45.5+4.9
Images and chartsGemini 3.1 Pro
- MMMU-ProGemini 3.1 Pro by 9.182.473.3+9.1
- LMArena · VisionGemini 3.1 Pro by 51 rating points12961245+51 rating
MathGemini 3.1 Pro
- FrontierMath Tier 4Gemini 3.1 Pro by 1726.89.8+17
- LiveBench · MathematicsGemini 3.1 Pro by 12.59178.5+12.5
- FrontierMath Tiers 1-3 (v2)Gemini 3.1 Pro by 8.559.651.2+8.5
Following instructionsGemini 3.1 Pro
- LiveBench · Instruction FollowingGemini 3.1 Pro by 19.379.159.8+19.3
- IFBenchGemini 3.1 Pro by 3.877.173.3+3.8
Other results16 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 3.1 Pro by 50.831.9-18.9+50.8
- ARC-AGI-1Gemini 3.1 Pro by 34.39863.7+34.3
- frontiermath_tier_4_v1Gemini 3.1 Pro by 14.616.72.1+14.6
- livebench_languageGemini 3.1 Pro by 14.485.471+14.4
- Artificial Analysis Coding IndexGemini 3.1 Pro by 12.768.856.1+12.7
- τ²-Bench Telecom (AA run)Gemini 3.1 Pro by 12.395.683.3+12.3
- AA Agentic IndexGPT-5.4 mini by 9.410.319.7+9.4
- Terminal-Bench 2.0Gemini 3.1 Pro by 8.568.560+8.5
- livebench_data_analysisGemini 3.1 Pro by 7.778.570.8+7.7
- ToolathlonGemini 3.1 Pro by 5.948.842.9+5.9
- AA IntelligenceGemini 3.1 Pro by 5.629.724.1+5.6
- vectara_avg_summary_lengthGemini 3.1 Pro by 53 rating points107.754.7+53 rating
- vectara_factual_consistencyGPT-5.4 mini by 4.989.694.5+4.9
- Finance Agent v2GPT-5.4 mini by 2.44345.4+2.4
- vectara_answer_ratetie99.4100tie
- Legal Agent Benchmarktie00tie
Questions people ask
Which is better, Gemini 3.1 Pro or GPT-5.4 mini?
Gemini 3.1 Pro wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. GPT-5.4 mini wins none, but costs 63% less.
Which is better for coding?
Gemini 3.1 Pro. It wins 6 of the 7 coding tests both models report; GPT-5.4 mini wins none, 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; GPT-5.4 mini costs $0.75 and $4.50. That makes GPT-5.4 mini about 63% 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 31 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.