Gemini 2.5 Pro vs Gemini 3 Pro
Wins 0 of 7 areas
—
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions
Gemini 3 Pro is the stronger all-rounder.Gemini 2.5 Pro is cheaper.
Scores updated · 54 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.
- Images and chartsUnderstanding pictures, charts and video06Gemini 3 Pro6 of 6 tests
- ReasoningHard problems that need careful thinking05Gemini 3 Pro5 of 5 tests
- CodingWriting and fixing software04Gemini 3 Pro4 of 4 tests
- AgentsCarrying out multi-step tasks on its own02Gemini 3 Pro2 of 2 tests
- Long documentsFinding answers in very long texts02Gemini 3 Pro2 of 2 tests
- Following instructionsDoing exactly what it is asked02Gemini 3 Pro2 of 2 tests
- FactsGetting facts right instead of making them up12Gemini 3 Pro2 of 4 tests · 1 tie
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.
Gemini 2.5 Pro 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 Pro pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+49.3points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+26.2points ahead
- Common-sense trick questionsSimpleBench+24.8points ahead
Where Gemini 2.5 Pro pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 54 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3 Pro
- LMArena · WebDevGemini 3 Pro by 216 rating points12241440+216 rating
- Terminal-Bench HardGemini 3 Pro by 15.226.541.7+15.2
- SWE-bench VerifiedGemini 3 Pro by 1363.276.2+13
- SciCodeGemini 3 Pro by 9.846.356.1+9.8
AgentsGemini 3 Pro
- BrowseCompGemini 3 Pro by 49.39.959.2+49.3
- GDPValGemini 3 Pro by 34.2034.2+34.2
ReasoningGemini 3 Pro
- ARC-AGI-2Gemini 3 Pro by 26.24.931.1+26.2
- SimpleBenchGemini 3 Pro by 24.851.676.4+24.8
- Humanity's Last ExamGemini 3 Pro by 21.71839.7+21.7
- GPQA DiamondGemini 3 Pro by 7.283.690.8+7.2
- CritPtGemini 3 Pro by 6.52.69.1+6.5
FactsGemini 3 Pro
- SimpleQA VerifiedGemini 3 Pro by 16.95672.9+16.9
- AA-Omniscience · AccuracyGemini 3 Pro by 16.73955.8+16.7
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 6.6713.6+6.6
- AA-Omniscience · Non-hallucinationtie9.18.5tie
Images and chartsGemini 3 Pro
- MathVistaGemini 3 Pro by 5.983.989.8+5.9
- MMMU-ProGemini 3 Pro by 5.374.980.2+5.3
- OCRBenchGemini 3 Pro by 4.485.990.3+4.4
- LMArena · VisionGemini 3 Pro by 42 rating points12631305+42 rating
- OCRBenchv2Gemini 3 Pro by 4.159.363.4+4.1
- Video-MMEGemini 3 Pro by 3.684.888.4+3.6
Long documentsGemini 3 Pro
- LongBench v2Gemini 3 Pro by 9.458.868.2+9.4
- AA-LCRGemini 3 Pro by 76976+7
Following instructionsGemini 3 Pro
- IFBenchGemini 3 Pro by 21.748.770.4+21.7
- Multi-ChallengeGemini 3 Pro by 12.153.665.7+12.1
Other results29 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 48.53.552+48.5
- τ²-BenchGemini 3 Pro by 445498+44
- ARC-AGI-1Gemini 3 Pro by 383775+38
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 3354.187.1+33
- AA-OmniscienceGemini 3 Pro by 31.6-16.415.3+31.6
- SimpleQAGemini 3 Pro by 21.350.872.1+21.3
- LiveCodeBenchGemini 3 Pro by 17.874.292+17.8
- HLE (with tools)Gemini 3 Pro by 17.428.445.8+17.4
- AIME 2025Gemini 3 Pro by 1783100+17
- HMMT Feb. 2025Gemini 3 Pro by 16.780.897.5+16.7
- frontiermath_tier_4_v1Gemini 3 Pro by 14.64.218.8+14.6
- HMMT Nov. 2025Gemini 3 Pro by 13.38093.3+13.3
- AA IntelligenceGemini 3 Pro by 131528+13
- MATH-VisionGemini 3 Pro by 12.873.386.1+12.8
- HarmBenchGemini 3 Pro by 7.165.472.5+7.1
- matharena_visual_math_overallGemini 3 Pro by 777.284.2+7
- vectara_factual_consistencyGemini 2.5 Pro by 6.69386.4+6.6
- MMLU-ProGemini 3 Pro by 4.18690.1+4.1
- VideoMMMUGemini 3 Pro by 483.687.6+4
- MRCRGemini 2.5 Pro by 3.39389.7+3.3
- anthropic_red_teamGemini 2.5 Pro by 2.499.597+2.4
- bbqGemini 3 Pro by 296.498.4+2
- OCRBench v2 (Chinese)Gemini 3 Pro by 1.662.263.8+1.6
- XSTestGemini 2.5 Pro by 1.498.797.3+1.4
- simple_safety_teststie9797.5tie
- vectara_avg_summary_lengthtie106.4101.9tie
- AIR-Bench 2024tie73.673.2tie
- vectara_answer_ratetie99.199.4tie
- Artificial Analysis Coding Indextie46.746.5tie
Questions people ask
Which is better, Gemini 2.5 Pro or Gemini 3 Pro?
Gemini 3 Pro wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, long documents and following instructions. Gemini 2.5 Pro wins none, but costs 20% less.
Which is better for coding?
Gemini 3 Pro. It wins 4 of the 4 coding tests both models report; Gemini 2.5 Pro wins none.
Which is cheaper?
Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; Gemini 3 Pro costs $2.00 and $12.00. That makes Gemini 2.5 Pro about 20% cheaper for the same work.
How do you compare the two?
We use the 54 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 29 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.