Gemini 3.1 Pro vs GPT-5
Wins 7 of 8 areas
Coding · 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 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 software80Gemini 3.1 Pro8 of 8 tests
- ReasoningHard problems that need careful thinking50Gemini 3.1 Pro5 of 5 tests
- FactsGetting facts right instead of making them up40Gemini 3.1 Pro4 of 4 tests
- Images and chartsUnderstanding pictures, charts and video40Gemini 3.1 Pro4 of 4 tests
- MathCompetition and research-level math30Gemini 3.1 Pro3 of 3 tests
- Following instructionsDoing exactly what it is asked20Gemini 3.1 Pro2 of 2 tests
- Long documentsFinding answers in very long texts10Gemini 3.1 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own11Even1 each · 1 tie
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 costs 20% 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 Gemini 3.1 Pro pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+67.2points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+38.6points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+31.3points ahead
Where GPT-5 pulls ahead
- Real work tasks from 44 professionsGDPVal+6.3points 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 38.673.835.2+38.6
- SWE-bench MultilingualGemini 3.1 Pro by 21.676.955.3+21.6
- Terminal-Bench HardGemini 3.1 Pro by 21.253.832.6+21.2
- SciCodeGemini 3.1 Pro by 15.858.742.9+15.8
- SWE-bench ProGemini 3.1 Pro by 12.454.241.8+12.4
- SWE-bench VerifiedGemini 3.1 Pro by 5.780.674.9+5.7
- LiveCodeBench v6Gemini 3.1 Pro by 4.791.787+4.7
- LMArena · WebDevGemini 3.1 Pro by 29 rating points14461417+29 rating
AgentsEven
- BrowseCompGemini 3.1 Pro by 3185.954.9+31
- GDPValGPT-5 by 6.314.721+6.3
- τ-Bench V3 · Bankingtie21.422.1tie
ReasoningGemini 3.1 Pro
- ARC-AGI-2Gemini 3.1 Pro by 67.277.19.9+67.2
- SimpleBenchGemini 3.1 Pro by 22.979.656.7+22.9
- Humanity's Last ExamGemini 3.1 Pro by 18.54728.5+18.5
- CritPtGemini 3.1 Pro by 1217.75.7+12
- GPQA DiamondGemini 3.1 Pro by 8.794.185.4+8.7
FactsGemini 3.1 Pro
- AA-Omniscience · Non-hallucinationGemini 3.1 Pro by 31.349.117.8+31.3
- SimpleQA VerifiedGemini 3.1 Pro by 23.473.550.1+23.4
- AA-Omniscience · AccuracyGemini 3.1 Pro by 14.654.940.3+14.6
- Vectara HHEM hallucination ratelower is betterGemini 3.1 Pro by 4.710.415.1+4.7
Images and chartsGemini 3.1 Pro
- LMArena · VisionGemini 3.1 Pro by 88 rating points12961208+88 rating
- MMMU-ProGemini 3.1 Pro by 8.282.474.2+8.2
- OCRBenchv2Gemini 3.1 Pro by 7.362.855.5+7.3
- CharXiv (RQ)Gemini 3.1 Pro by 2.283.381.1+2.2
MathGemini 3.1 Pro
- IMOAnswerBenchGemini 3.1 Pro by 58176+5
- FrontierMath Tier 4Gemini 3.1 Pro by 4.826.821.9+4.8
- FrontierMath Tiers 1-3 (v2)Gemini 3.1 Pro by 4.359.655.4+4.3
Following instructionsGemini 3.1 Pro
- IFBenchGemini 3.1 Pro by 477.173.1+4
- Multi-ChallengeGemini 3.1 Pro by 1.871.469.6+1.8
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.
- AA-OmniscienceGemini 3.1 Pro by 40.631.9-8.7+40.6
- Terminal-Bench 2.0Gemini 3.1 Pro by 33.368.535.2+33.3
- ARC-AGI-1Gemini 3.1 Pro by 32.39865.7+32.3
- Artificial Analysis Coding IndexGemini 3.1 Pro by 3168.837.8+31
- AA Agentic IndexGPT-5 by 16.210.326.5+16.2
- HLE (with tools)Gemini 3.1 Pro by 16.251.435.2+16.2
- τ²-Bench Telecom (AA run)Gemini 3.1 Pro by 10.895.684.8+10.8
- matharena_visual_math_overallGemini 3.1 Pro by 10.689.478.8+10.6
- LiveCodeBenchGemini 3.1 Pro by 8.291.783.5+8.2
- AA IntelligenceGemini 3.1 Pro by 6.729.723+6.7
- HMMT Nov. 2025Gemini 3.1 Pro by 5.694.889.2+5.6
- vectara_avg_summary_lengthGPT-5 by 55 rating points107.7162.7+55 rating
- ERQAGemini 3.1 Pro by 5.170.865.7+5.1
- vectara_factual_consistencyGemini 3.1 Pro by 4.789.684.9+4.7
- frontiermath_tier_4_v1Gemini 3.1 Pro by 4.216.712.5+4.2
- MMLU-ProGemini 3.1 Pro by 3.99187.1+3.9
- vectara_answer_ratetie99.499.9tie
Questions people ask
Which is better, Gemini 3.1 Pro or GPT-5?
Gemini 3.1 Pro wins seven of the eight areas we test: coding, reasoning, facts, images and charts, math, long documents and following instructions. GPT-5 wins none, but costs 20% less. They are level on agents.
Which is better for coding?
Gemini 3.1 Pro. It wins 8 of the 8 coding tests both models report; GPT-5 wins none.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; GPT-5 costs $1.25 and $10.00. That makes GPT-5 about 20% 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.