Gemini 3 Pro vs GPT-5.1
Wins 4 of 7 areas
Coding · Agents · Reasoning · Images and charts
Wins 1 of 7 areas
Long documents
Gemini 3 Pro wins more areas, narrowly.GPT-5.1 is cheaper and better at long documents.
Scores updated · 49 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.
- ReasoningHard problems that need careful thinking50Gemini 3 Pro5 of 5 tests
- AgentsCarrying out multi-step tasks on its own20Gemini 3 Pro2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video20Gemini 3 Pro2 of 2 tests
- CodingWriting and fixing software21Gemini 3 Pro2 of 5 tests · 2 ties
- Long documentsFinding answers in very long texts01GPT-5.11 of 1 test
- FactsGetting facts right instead of making them up22Even2 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.
GPT-5.1 costs 20% 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
- Short factual questions, answered correctlySimpleQA Verified+24.9points ahead
- Common-sense trick questionsSimpleBench+23.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+18.1points ahead
Where GPT-5.1 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+39.6points ahead
- Reasons across sets of long documentsAA-LCR+4points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+3.8points ahead
Every test, side by side
All 49 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 12.856.143.3+12.8
- LMArena · WebDevGemini 3 Pro by 44 rating points14391395+44 rating
- Terminal-Bench HardGPT-5.1 by 3.841.745.5+3.8
- LiveCodeBench v6tie87.487tie
- SWE-bench Verifiedtie76.276.3tie
AgentsGemini 3 Pro
- GDPValGemini 3 Pro by 17.734.216.5+17.7
- BrowseCompGemini 3 Pro by 8.459.250.8+8.4
ReasoningGemini 3 Pro
- SimpleBenchGemini 3 Pro by 23.276.453.2+23.2
- ARC-AGI-2Gemini 3 Pro by 13.531.117.6+13.5
- Humanity's Last ExamGemini 3 Pro by 11.239.728.5+11.2
- CritPtGemini 3 Pro by 4.29.14.9+4.2
- GPQA DiamondGemini 3 Pro by 3.590.887.3+3.5
FactsEven
- AA-Omniscience · Non-hallucinationGPT-5.1 by 39.68.548.1+39.6
- SimpleQA VerifiedGemini 3 Pro by 24.972.948+24.9
- AA-Omniscience · AccuracyGemini 3 Pro by 18.155.837.7+18.1
- Vectara HHEM hallucination ratelower is betterGPT-5.1 by 1.513.612.1+1.5
Images and chartsGemini 3 Pro
- LMArena · VisionGemini 3 Pro by 55 rating points13051250+55 rating
- MMMU-ProGemini 3 Pro by 4.780.275.5+4.7
Following instructionsEven
- IFBenchGPT-5.1 by 2.570.472.9+2.5
- Multi-ChallengeGemini 3 Pro by 2.365.763.4+2.3
Other results28 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 Agentic IndexGemini 3 Pro by 30.45221.6+30.4
- putnam_2025Gemini 3 Pro by 27.575.848.3+27.5
- HarmBenchGPT-5.1 by 25.172.597.6+25.1
- τ²-BenchGemini 3 Pro by 15.39882.7+15.3
- vectara_avg_summary_lengthGPT-5.1 by 152.5 rating points101.9254.4+152.5 rating
- ARC-AGI-2 (Verified)Gemini 3 Pro by 13.531.117.6+13.5
- AIR-Bench 2024GPT-5.1 by 1373.286.2+13
- AA-OmniscienceGemini 3 Pro by 9.915.35.4+9.9
- bbqGemini 3 Pro by 9.798.488.7+9.7
- τ²-Bench (Retail)Gemini 3 Pro by 7.485.377.9+7.4
- matharena_visual_math_overallGemini 3 Pro by 7.384.276.9+7.3
- Terminal-Bench 2.0Gemini 3 Pro by 6.654.247.6+6.6
- frontiermath_tier_4_v1Gemini 3 Pro by 6.318.812.5+6.3
- AIME 2025Gemini 3 Pro by 610094+6
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 5.287.181.9+5.2
- AA IntelligenceGemini 3 Pro by 3.32824.7+3.3
- HLE (with tools)Gemini 3 Pro by 3.145.842.7+3.1
- MMLU-ProGemini 3 Pro by 3.190.187+3.1
- Artificial Analysis Coding IndexGPT-5.1 by 2.946.549.4+2.9
- anthropic_red_teamGPT-5.1 by 2.49799.5+2.4
- simple_safety_testsGPT-5.1 by 2.397.599.8+2.3
- ARC-AGI-1Gemini 3 Pro by 2.27572.8+2.2
- HMMT Nov. 2025Gemini 3 Pro by 1.693.391.7+1.6
- vectara_factual_consistencyGPT-5.1 by 1.586.487.9+1.5
- HMMT Feb. 2025Gemini 3 Pro by 1.297.596.3+1.2
- XSTesttie97.398.2tie
- MMMLUtie91.891tie
- vectara_answer_ratetie99.4100tie
Questions people ask
Which is better, Gemini 3 Pro or GPT-5.1?
Gemini 3 Pro wins four of the seven areas where both have results: coding, agents, reasoning and images and charts. GPT-5.1 wins long documents, and costs 20% less. They are level on facts and following instructions.
Which is better for coding?
Gemini 3 Pro. It wins 2 of the 5 coding tests both models report; GPT-5.1 wins 1, and 2 are ties.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; GPT-5.1 costs $1.25 and $10.00. That makes GPT-5.1 about 20% cheaper for the same work.
How do you compare the two?
We use the 49 benchmark tests both models have published scores on. The verdict counts the 21 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 28 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.