Gemini 3.1 Pro vs Gemini 3.5 Flash
Wins 5 of 8 areas
Reasoning · Facts · Math · Long documents · Following instructions
Wins 3 of 8 areas
Coding · Agents · Images and charts
Gemini 3.1 Pro is the stronger all-rounder.Gemini 3.5 Flash is cheaper and better at agents.
Both rank among the ten best models we track in following instructions.
Scores updated · 73 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.
- ReasoningHard problems that need careful thinking60Gemini 3.1 Pro6 of 6 tests
- FactsGetting facts right instead of making them up30Gemini 3.1 Pro3 of 3 tests
- MathCompetition and research-level math21Gemini 3.1 Pro2 of 5 tests · 2 ties
- Long documentsFinding answers in very long texts10Gemini 3.1 Pro1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked10Gemini 3.1 Pro1 of 2 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own07Gemini 3.5 Flash7 of 7 tests
- CodingWriting and fixing software24Gemini 3.5 Flash4 of 7 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video02Gemini 3.5 Flash2 of 4 tests · 2 ties
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 3.5 Flash costs 25% 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
- Hard command-line tasks in a real terminalTerminal-Bench Hard+12.9points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+11.3points ahead
- Reasons across sets of long documentsAA-LCR+8.7points ahead
Where Gemini 3.5 Flash pulls ahead
- Real work tasks from 44 professionsGDPVal+20.5points ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+15.1points ahead
- Multi-step tasks using many real software toolsMCP Atlas+14.4points ahead
Every test, side by side
All 73 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3.5 Flash
- Terminal-Bench HardGemini 3.1 Pro by 12.953.840.9+12.9
- LMArena · WebDevGemini 3.5 Flash by 53 rating points14461499+53 rating
- LiveBench · Agentic CodingGemini 3.5 Flash by 4.944.149+4.9
- Terminal-Bench 2.1Gemini 3.5 Flash by 4.973.878.7+4.9
- SciCodeGemini 3.1 Pro by 4.858.753.9+4.8
- LiveBench · CodingGemini 3.5 Flash by 1.776.578.2+1.7
- SWE-bench Protie54.255.1tie
AgentsGemini 3.5 Flash
- GDPValGemini 3.5 Flash by 20.514.735.2+20.5
- AA ApexAgentsGemini 3.5 Flash by 15.13247.1+15.1
- MCP AtlasGemini 3.5 Flash by 14.469.283.6+14.4
- τ-Bench V3 · BankingGemini 3.5 Flash by 10.821.432.2+10.8
- AA IT-Bench SREGemini 3.5 Flash by 1030.340.3+10
- Terminal-Bench 4.0Gemini 3.5 Flash by 2.646.6+2.6
- OSWorld-VerifiedGemini 3.5 Flash by 2.276.278.4+2.2
ReasoningGemini 3.1 Pro
- ARC-AGI-2Gemini 3.1 Pro by 577.172.1+5
- CritPtGemini 3.1 Pro by 4.617.713.1+4.6
- Humanity's Last ExamGemini 3.1 Pro by 4.34742.7+4.3
- SimpleBenchGemini 3.1 Pro by 2.979.676.7+2.9
- LiveBench · ReasoningGemini 3.1 Pro by 28482+2
- GPQA DiamondGemini 3.1 Pro by 1.994.192.2+1.9
FactsGemini 3.1 Pro
- AA-Omniscience · Non-hallucinationGemini 3.1 Pro by 11.349.137.8+11.3
- SimpleQA VerifiedGemini 3.1 Pro by 7.373.566.2+7.3
- AA-Omniscience · AccuracyGemini 3.1 Pro by 3.554.951.4+3.5
Images and chartsGemini 3.5 Flash
- MMMU-ProGemini 3.5 Flash by 1.982.484.3+1.9
- LMArena · VisionGemini 3.5 Flash by 14 rating points12961310+14 rating
- CharXiv (RQ)tie83.384.2tie
- Video-MMEtie86.787.2tie
MathGemini 3.1 Pro
- AIME 2026Gemini 3.1 Pro by 3.398.395+3.3
- FrontierMath Tiers 1-3 (v2)Gemini 3.5 Flash by 3.159.662.8+3.1
- LiveBench · MathematicsGemini 3.1 Pro by 2.89188.2+2.8
- HMMT Feb. 2026tie94.795.5tie
- FrontierMath Tier 4tie26.826.8tie
Long documentsGemini 3.1 Pro
- AA-LCRGemini 3.1 Pro by 8.78273.3+8.7
- MRCR v2 8 needle 128k (average)tie26.326.6tie
Following instructionsGemini 3.1 Pro
- LiveBench · Instruction FollowingGemini 3.1 Pro by 3.579.175.6+3.5
- IFBenchtie77.176.3tie
Other results37 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 3.5 Flash by 387 rating points9621349+387 rating
- GDPval-AA v2 EloGemini 3.5 Flash by 384 rating points9651349+384 rating
- GDPval-AA (Elo)Gemini 3.5 Flash by 339 rating points13171656+339 rating
- DeepSWEGemini 3.5 Flash by 271037+27
- DeepSWE 1.1Gemini 3.5 Flash by 251237+25
- AA Agentic IndexGemini 3.5 Flash by 1710.327.3+17
- Finance Agent v2Gemini 3.5 Flash by 14.94357.9+14.9
- livebench_data_analysisGemini 3.1 Pro by 13.678.564.9+13.6
- TOMATOGemini 3.5 Flash by 11.560.471.9+11.5
- AA-OmniscienceGemini 3.1 Pro by 10.731.921.2+10.7
- LVBenchGemini 3.5 Flash by 10.166.276.3+10.1
- Terminal-Bench 2.0Gemini 3.5 Flash by 7.768.576.2+7.7
- ToolathlonGemini 3.5 Flash by 7.748.856.5+7.7
- HiL-Bench (Tools-allowed)Gemini 3.1 Pro by 7.635.327.7+7.6
- Blueprint-Bench 2Gemini 3.5 Flash by 7.126.533.6+7.1
- MLE-BenchGemini 3.5 Flash by 7.142.649.7+7.1
- VideoSimpleQAGemini 3.5 Flash by 67076+6
- ARC-AGI-1Gemini 3.1 Pro by 5.59892.5+5.5
- TVBenchGemini 3.5 Flash by 5.47176.4+5.4
- MinervaGemini 3.5 Flash by 5.163.568.6+5.1
- LiveBenchGemini 3.1 Pro by 4.979.975+4.9
- Humanity’s Last ExamGemini 3.1 Pro by 4.244.440.2+4.2
- mrcr_v2_8needle_1m_pointwiseGemini 3.1 Pro by 4.226.322.1+4.2
- AA IntelligenceGemini 3.5 Flash by 2.929.732.6+2.9
- OVBenchGemini 3.1 Pro by 2.358.856.5+2.3
- frontiermath_tier_4_v1Gemini 3.1 Pro by 2.116.714.6+2.1
- Finance AgentGemini 3.1 Pro by 1.859.757.9+1.8
- CharXiv Reasoning (With tools)Gemini 3.5 Flash by 1.783.284.9+1.7
- Artificial Analysis Coding IndexGemini 3.5 Flash by 1.368.870.1+1.3
- VideoHolmesGemini 3.5 Flash by 1.265.967.1+1.2
- LongVideoBenchGemini 3.5 Flash by 1.176.577.6+1.1
- Legal Agent Benchmarktie00.8tie
- livebench_languagetie85.484.6tie
- MotionBenchtie69.970.6tie
- matharena_visual_math_overalltie89.489.9tie
- GDM-MRCR v2 (8-needle) (1M (pointwise))tie26.326.6tie
- τ²-Bench Telecom (AA run)tie95.695.3tie
Questions people ask
Which is better, Gemini 3.1 Pro or Gemini 3.5 Flash?
Gemini 3.1 Pro wins five of the eight areas we test: reasoning, facts, math, long documents and following instructions. Gemini 3.5 Flash wins coding, agents and images and charts, and costs 25% less.
Which is better for coding?
Gemini 3.5 Flash. It wins 4 of the 7 coding tests both models report; Gemini 3.1 Pro wins 2, 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 3.5 Flash costs $1.50 and $9.00. That makes Gemini 3.5 Flash about 25% cheaper for the same work.
How do you compare the two?
We use the 73 benchmark tests both models have published scores on. The verdict counts the 36 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 37 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.