Gemini 3.1 Pro vs Gemini 3.6 Flash
Wins 4 of 8 areas
Reasoning · Facts · Math · Following instructions
Wins 3 of 8 areas
Coding · Agents · Images and charts
Gemini 3.1 Pro wins more areas, narrowly.Gemini 3.6 Flash is cheaper and better at agents.
Scores updated · 46 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.
- MathCompetition and research-level math40Gemini 3.1 Pro4 of 5 tests · 1 tie
- ReasoningHard problems that need careful thinking41Gemini 3.1 Pro4 of 5 tests
- FactsGetting facts right instead of making them up30Gemini 3.1 Pro3 of 3 tests
- Following instructionsDoing exactly what it is asked10Gemini 3.1 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own04Gemini 3.6 Flash4 of 4 tests
- CodingWriting and fixing software14Gemini 3.6 Flash4 of 6 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video01Gemini 3.6 Flash1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts11Even1 each
Following instructions 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.
Gemini 3.6 Flash costs 68% 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+16.7points ahead
- Short factual questions, answered correctlySimpleQA Verified+7.3points ahead
- Research-level physics problemsCritPt+7.1points ahead
Where Gemini 3.6 Flash pulls ahead
- Finds one of eight look-alike replies in a long chatMRCR v2 8 needle 128k (average)+65.5points ahead
- Real work tasks from 44 professionsGDPVal+24.6points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+8.5points ahead
Every test, side by side
All 46 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3.6 Flash
- LMArena · WebDevGemini 3.6 Flash by 90 rating points14461536+90 rating
- SciCodeGemini 3.1 Pro by 5.358.753.4+5.3
- SWE-bench ProGemini 3.6 Flash by 4.554.258.7+4.5
- Terminal-Bench 2.1Gemini 3.6 Flash by 3.773.877.5+3.7
- LiveBench · CodingGemini 3.6 Flash by 1.476.577.9+1.4
- LiveBench · Agentic Codingtie44.143.4tie
AgentsGemini 3.6 Flash
- GDPValGemini 3.6 Flash by 24.614.739.3+24.6
- τ-Bench V3 · BankingGemini 3.6 Flash by 8.521.429.9+8.5
- OSWorld-VerifiedGemini 3.6 Flash by 6.876.283+6.8
- Terminal-Bench 4.0Gemini 3.6 Flash by 3.147.1+3.1
ReasoningGemini 3.1 Pro
- ARC-AGI-2Gemini 3.1 Pro by 16.777.160.4+16.7
- CritPtGemini 3.1 Pro by 7.117.710.6+7.1
- Humanity's Last ExamGemini 3.1 Pro by 6.24740.8+6.2
- GPQA DiamondGemini 3.1 Pro by 1.394.192.8+1.3
- LiveBench · ReasoningGemini 3.6 Flash by 1.28485.2+1.2
FactsGemini 3.1 Pro
- SimpleQA VerifiedGemini 3.1 Pro by 7.373.566.2+7.3
- AA-Omniscience · AccuracyGemini 3.1 Pro by 4.954.950+4.9
- AA-Omniscience · Non-hallucinationGemini 3.1 Pro by 4.749.144.4+4.7
Images and chartsGemini 3.6 Flash
- CharXiv (RQ)Gemini 3.6 Flash by 6.183.389.4+6.1
- MMMU-Protie82.483.2tie
- LMArena · Visiontie12961297tie
MathGemini 3.1 Pro
- HMMT Feb. 2026Gemini 3.1 Pro by 5.394.789.4+5.3
- FrontierMath Tier 4Gemini 3.1 Pro by 4.826.821.9+4.8
- LiveBench · MathematicsGemini 3.1 Pro by 4.69186.4+4.6
- AIME 2026Gemini 3.1 Pro by 1.698.396.7+1.6
- FrontierMath Tiers 1-3 (v2)tie59.659tie
Long documentsEven
- MRCR v2 8 needle 128k (average)Gemini 3.6 Flash by 65.526.391.8+65.5
- AA-LCRGemini 3.1 Pro by 28280+2
Following instructionsGemini 3.1 Pro
- LiveBench · Instruction FollowingGemini 3.1 Pro by 3.779.175.4+3.7
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.
- GDPval-AA v2Gemini 3.6 Flash by 459 rating points9621421+459 rating
- GDPval-AA v2 EloGemini 3.6 Flash by 457 rating points9651422+457 rating
- DeepSWEGemini 3.6 Flash by 391049+39
- DeepSWE 1.1Gemini 3.6 Flash by 371249+37
- GDM-MRCR v2 (8-needle) (1M (pointwise))Gemini 3.6 Flash by 27.726.354+27.7
- MLE-BenchGemini 3.6 Flash by 21.342.663.9+21.3
- AA Agentic IndexGemini 3.6 Flash by 19.910.330.2+19.9
- LVBenchGemini 3.6 Flash by 1866.284.2+18
- livebench_data_analysisGemini 3.1 Pro by 15.578.563+15.5
- AA-OmniscienceGemini 3.1 Pro by 9.831.922.1+9.8
- ARC-AGI-1Gemini 3.1 Pro by 6.89891.2+6.8
- CharXiv Reasoning (With tools)Gemini 3.6 Flash by 6.283.289.4+6.2
- GDP (Surge AI)Gemini 3.6 Flash by 5.316.722+5.3
- AA IntelligenceGemini 3.6 Flash by 4.329.734+4.3
- livebench_languageGemini 3.1 Pro by 1.585.483.9+1.5
- matharena_visual_math_overallGemini 3.1 Pro by 1.589.487.8+1.5
- Artificial Analysis Coding Indextie68.869.2tie
Questions people ask
Which is better, Gemini 3.1 Pro or Gemini 3.6 Flash?
Gemini 3.1 Pro wins four of the eight areas we test: reasoning, facts, math and following instructions. Gemini 3.6 Flash wins coding, agents and images and charts, and costs 68% less. They are level on long documents.
Which is better for coding?
Gemini 3.6 Flash. It wins 4 of the 6 coding tests both models report; Gemini 3.1 Pro wins 1, 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; Gemini 3.6 Flash costs $0.75 and $3.75. That makes Gemini 3.6 Flash about 68% cheaper for the same work.
How do you compare the two?
We use the 46 benchmark tests both models have published scores on. The verdict counts the 29 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.