The Model Index
Up to four models

Gemini 3.7 Flash vs Gemini 3.6 Flash

On the index (GPT-4 = 100), Gemini 3.7 Flash scores 170 and Gemini 3.6 Flash 161. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Gemini 3.7 Flash
Presets
Gemini 3.7 Flash
170154–186
WeightsClosed
ReleasedAug 2026
Price in / out$0.75 / $3.75
Context1.05M
Core tests taken6 of 9
Gemini 3.6 Flash
161145–176
WeightsClosed
ReleasedJul 2026
Price in / out$0.75 / $3.75
Context1.05M
Core tests taken5 of 9

Gemini 3.7 Flash leads Gemini 3.6 Flash by 10 points, ahead on 5 of 5 shared core tests, at a similar price.

Test by testthird-party results

Gemini 3.7 FlashGemini 3.6 Flash
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)GPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMMMU ProMMMU ProSimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)Chess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)CODINGCode MigrationCode MigrationCyberBench v1.1CyberBench v1.1IOIIOILiveCodeBenchLiveCodeBenchProgramBenchProgramBenchSWE-benchSWE-benchVibe Code Bench v1.1Vibe Code Bench v1.1DeepSWEDeepSWEMirrorCodeMirrorCodeAGENTSAPEX-AgentsAPEX-AgentsTerminal-Bench-ScienceTerminal-Bench-ScienceDTBenchDTBenchFurniture assemblyFurniture assemblyLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)EMBEMBLegalBenchLegalBenchLegal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribeMortgageTaxMortgageTaxTax Agent BenchTax Agent BenchTaxEval v2TaxEval v2NOVEL PROBLEMSARC-AGI-2ARC-AGI-2ARC-AGI-1ARC-AGI-1
Every result as a table
TestGemini 3.7 FlashGemini 3.6 Flash
Reasoning & knowledge
GPQA Diamond (Epoch AI)94.8%94.1%
GPQA Diamond (Vals AI)93.9%93.4%
MMLU Pro90.1%89.3%
MMMU Pro89.0%88.4%
SimpleQA Verified69.2%66.2%
Maths
FrontierMath (tiers 1-3)71.6%59.0%
AIME-style maths (OTIS mock)97.2%94.2%
Chess puzzles66.0%42.0%
FrontierMath (tier 4)36.6%21.9%
Coding
Code Migration34.8%30.9%
CyberBench v1.143.8%45.4%
IOI67.8%35.1%
LiveCodeBench88.6%88.1%
ProgramBench0.0%0.0%
SWE-bench80.8%79.6%
Vibe Code Bench v1.170.4%64.0%
DeepSWE65.5%46.7%
MirrorCode41.1%23.3%
Agents
APEX-Agents67.8%46.9%
Terminal-Bench-Science7.1%4.3%
DTBench96.8%95.5%
Furniture assembly26.7%23.3%
LMCA50.4%44.9%
Mystery game puzzles37.0%30.0%
Professional work
Finance Agent (v2)59.0%56.3%
EMB71.3%65.4%
LegalBench87.3%86.7%
Legal Research Bench34.6%25.0%
MedCode53.4%53.1%
MedScribe83.9%79.7%
MortgageTax66.6%67.6%
Tax Agent Bench22.8%18.0%
TaxEval v274.7%74.9%
Novel problems
ARC-AGI-284.6%60.4%
ARC-AGI-195.5%91.2%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG