Gemini 3.5 Flash vs Gemini 3.6 Flash
Wins 3 of 8 areas
Reasoning · Images and charts · Math
Wins 2 of 8 areas
Agents · Long documents
Gemini 3.5 Flash wins more areas, narrowly.Gemini 3.6 Flash is cheaper and better at long documents.
Scores updated · 45 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 math41Gemini 3.5 Flash4 of 5 tests
- ReasoningHard problems that need careful thinking31Gemini 3.5 Flash3 of 5 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video21Gemini 3.5 Flash2 of 3 tests
- Long documentsFinding answers in very long texts02Gemini 3.6 Flash2 of 2 tests
- AgentsCarrying out multi-step tasks on its own12Gemini 3.6 Flash2 of 4 tests · 1 tie
- CodingWriting and fixing software22Even2 each · 2 ties
- FactsGetting facts right instead of making them up11Even1 each · 1 tie
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
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 57% 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 3.5 Flash pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+11.7points ahead
- Harvard-MIT high-school math contest problemsHMMT Feb. 2026+6.1points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+5.6points ahead
Where Gemini 3.6 Flash pulls ahead
- Finds one of eight look-alike replies in a long chatMRCR v2 8 needle 128k (average)+65.2points ahead
- Reasons across sets of long documentsAA-LCR+6.7points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+6.6points ahead
Every test, side by side
All 45 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- LiveBench · Agentic CodingGemini 3.5 Flash by 5.64943.4+5.6
- LMArena · WebDevGemini 3.6 Flash by 39 rating points14991538+39 rating
- SWE-bench ProGemini 3.6 Flash by 3.655.158.7+3.6
- Terminal-Bench 2.1Gemini 3.5 Flash by 1.278.777.5+1.2
- SciCodetie53.953.4tie
- LiveBench · Codingtie78.277.9tie
AgentsGemini 3.6 Flash
- OSWorld-VerifiedGemini 3.6 Flash by 4.678.483+4.6
- GDPValGemini 3.6 Flash by 4.135.239.3+4.1
- τ-Bench V3 · BankingGemini 3.5 Flash by 2.332.229.9+2.3
- Terminal-Bench 4.0tie6.67.1tie
ReasoningGemini 3.5 Flash
- ARC-AGI-2Gemini 3.5 Flash by 11.772.160.4+11.7
- LiveBench · ReasoningGemini 3.6 Flash by 3.28285.2+3.2
- CritPtGemini 3.5 Flash by 2.513.110.6+2.5
- Humanity's Last ExamGemini 3.5 Flash by 1.942.740.8+1.9
- GPQA Diamondtie92.292.8tie
FactsEven
- AA-Omniscience · Non-hallucinationGemini 3.6 Flash by 6.637.844.4+6.6
- AA-Omniscience · AccuracyGemini 3.5 Flash by 1.451.450+1.4
- SimpleQA Verifiedtie66.266.2tie
Images and chartsGemini 3.5 Flash
- CharXiv (RQ)Gemini 3.6 Flash by 5.284.289.4+5.2
- LMArena · VisionGemini 3.5 Flash by 13 rating points13101297+13 rating
- MMMU-ProGemini 3.5 Flash by 1.184.383.2+1.1
MathGemini 3.5 Flash
- HMMT Feb. 2026Gemini 3.5 Flash by 6.195.589.4+6.1
- FrontierMath Tier 4Gemini 3.5 Flash by 4.826.821.9+4.8
- FrontierMath Tiers 1-3 (v2)Gemini 3.5 Flash by 3.862.859+3.8
- LiveBench · MathematicsGemini 3.5 Flash by 1.888.286.4+1.8
- AIME 2026Gemini 3.6 Flash by 1.79596.7+1.7
Long documentsGemini 3.6 Flash
- MRCR v2 8 needle 128k (average)Gemini 3.6 Flash by 65.226.691.8+65.2
- AA-LCRGemini 3.6 Flash by 6.773.380+6.7
Following instructionsEven
- LiveBench · Instruction Followingtie75.675.4tie
Other results16 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.
- GDM-MRCR v2 (8-needle) (1M (pointwise))Gemini 3.6 Flash by 27.426.654+27.4
- MLE-BenchGemini 3.6 Flash by 14.249.763.9+14.2
- DeepSWEGemini 3.6 Flash by 123749+12
- DeepSWE 1.1Gemini 3.6 Flash by 123749+12
- LVBenchGemini 3.6 Flash by 7.976.384.2+7.9
- GDPval-AA v2 EloGemini 3.6 Flash by 73 rating points13491422+73 rating
- GDPval-AA v2Gemini 3.6 Flash by 72 rating points13491421+72 rating
- CharXiv Reasoning (With tools)Gemini 3.6 Flash by 4.584.989.4+4.5
- AA Agentic IndexGemini 3.6 Flash by 2.927.330.2+2.9
- matharena_visual_math_overallGemini 3.5 Flash by 289.987.8+2
- livebench_data_analysisGemini 3.5 Flash by 1.964.963+1.9
- AA IntelligenceGemini 3.6 Flash by 1.432.634+1.4
- ARC-AGI-1Gemini 3.5 Flash by 1.392.591.2+1.3
- AA-Omnisciencetie21.222.1tie
- Artificial Analysis Coding Indextie70.169.2tie
- livebench_languagetie84.683.9tie
Questions people ask
Which is better, Gemini 3.5 Flash or Gemini 3.6 Flash?
Gemini 3.5 Flash wins three of the eight areas we test: reasoning, images and charts and math. Gemini 3.6 Flash wins agents and long documents, and costs 57% less. They are level on coding, facts and following instructions.
Which is better for coding?
Neither. They win 2 coding tests each of the 6 both models report, and 2 are ties.
Which is cheaper?
Gemini 3.5 Flash costs $1.50 per million input tokens and $9.00 per million output tokens; Gemini 3.6 Flash costs $0.75 and $3.75. That makes Gemini 3.6 Flash about 57% cheaper for the same work.
How do you compare the two?
We use the 45 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 16 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.