Gemini 2.5 Pro vs Gemini 3.1 Flash Lite Preview
Wins 3 of 8 areas
Reasoning · Facts · Images and charts
Wins 3 of 8 areas
Math · Long documents · Following instructions
The two are evenly matched.Gemini 2.5 Pro is better at reasoning and facts; Gemini 3.1 Flash Lite Preview at following instructions and math.
Scores updated · 29 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 thinking20Gemini 2.5 Pro2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up21Gemini 2.5 Pro2 of 3 tests
- Images and chartsUnderstanding pictures, charts and video10Gemini 2.5 Pro1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked02Gemini 3.1 Flash Lite Preview2 of 2 tests
- MathCompetition and research-level math01Gemini 3.1 Flash Lite Preview1 of 1 test
- Long documentsFinding answers in very long texts01Gemini 3.1 Flash Lite Preview1 of 1 test
- CodingWriting and fixing software22Even2 each
- AgentsCarrying out multi-step tasks on its own00Even0 each · 3 ties
Math and long documents rest on a single test each.
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 3.1 Flash Lite Preview costs 84% 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 2.5 Pro pulls ahead
- Code for real scientific research problemsSciCode+2.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+2.8points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+2.3points ahead
Where Gemini 3.1 Flash Lite Preview pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+28.5points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+8.2points ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+7points ahead
Every test, side by side
All 29 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- LMArena · WebDevGemini 3.1 Flash Lite Preview by 30 rating points12241254+30 rating
- SciCodeGemini 2.5 Pro by 2.946.343.4+2.9
- Terminal-Bench 2.1Gemini 3.1 Flash Lite Preview by 2.628.531.1+2.6
- Terminal-Bench HardGemini 2.5 Pro by 2.326.524.2+2.3
AgentsEven
- Terminal-Bench 4.0tie00.5tie
- GDPValtie00tie
- τ-Bench V3 · Bankingtie9.79.7tie
ReasoningGemini 2.5 Pro
- CritPtGemini 2.5 Pro by 1.52.61.1+1.5
- GPQA DiamondGemini 2.5 Pro by 1.483.682.2+1.4
- Humanity's Last Examtie1817.2tie
FactsGemini 2.5 Pro
- AA-Omniscience · Non-hallucinationGemini 3.1 Flash Lite Preview by 8.29.117.3+8.2
- AA-Omniscience · AccuracyGemini 2.5 Pro by 2.83936.3+2.8
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 1.278.2+1.2
Images and chartsGemini 2.5 Pro
- LMArena · VisionGemini 2.5 Pro by 23 rating points12631240+23 rating
- MMMU-Protie74.975.5tie
MathGemini 3.1 Flash Lite Preview
- FrontierMath Tiers 1-3 (v2)Gemini 3.1 Flash Lite Preview by 3.124.627.7+3.1
Long documentsGemini 3.1 Flash Lite Preview
- AA-LCRGemini 3.1 Flash Lite Preview by 5.36974.3+5.3
Following instructionsGemini 3.1 Flash Lite Preview
- IFBenchGemini 3.1 Flash Lite Preview by 28.548.777.2+28.5
- Multi-ChallengeGemini 3.1 Flash Lite Preview by 753.660.6+7
Other results10 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.
- τ²-Bench Telecom (AA run)Gemini 2.5 Pro by 22.854.131.3+22.8
- Artificial Analysis Coding IndexGemini 2.5 Pro by 1246.734.7+12
- SimpleQAGemini 2.5 Pro by 7.550.843.3+7.5
- vectara_avg_summary_lengthGemini 2.5 Pro by 43.8 rating points106.462.6+43.8 rating
- vectara_factual_consistencyGemini 2.5 Pro by 1.29391.8+1.2
- VideoMMMUGemini 3.1 Flash Lite Preview by 1.283.684.8+1.2
- AA Intelligencetie1515.6tie
- vectara_answer_ratetie99.199.6tie
- AA Agentic Indextie3.53.2tie
- AA-Omnisciencetie-16.4-16.4tie
Questions people ask
Which is better, Gemini 2.5 Pro or Gemini 3.1 Flash Lite Preview?
Gemini 2.5 Pro and Gemini 3.1 Flash Lite Preview each win three of the eight areas we test. Gemini 2.5 Pro wins reasoning, facts and images and charts; Gemini 3.1 Flash Lite Preview wins math, long documents and following instructions. Gemini 3.1 Flash Lite Preview costs 84% less. They are level on coding and agents.
Which is better for coding?
Neither. They win 2 coding tests each of the 4 both models report.
Which is cheaper?
Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; Gemini 3.1 Flash Lite Preview costs $0.25 and $1.50. That makes Gemini 3.1 Flash Lite Preview about 84% cheaper for the same work.
How do you compare the two?
We use the 29 benchmark tests both models have published scores on. The verdict counts the 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 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.