DataLearner logo

Gemini 3.6 FlashvsGemini 3.5 Flash

Across 13 shared benchmarks, Gemini 3.6 Flash leads overall: Gemini 3.6 Flash wins 11, Gemini 3.5 Flash wins 2, with 0 ties and an average score difference of +15.58.

Google Deep Mind
Gemini 3.6 Flash

Google Deep Mind · 2026-07-21 · Multimodal model

Google Deep Mind
Gemini 3.5 Flash

Google Deep Mind · 2026-06-20 · Multimodal model

Gemini 3.6 Flash11 wins(85%)(15%)2 winsGemini 3.5 Flash

Benchmark scores

Grouped by capability, sorted by largest gap within each. 13 shared benchmarks.

AI Agent - Tool Usage

Gemini 3.6 Flash 3/3
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
MLE-Bench63.901 / 3Thinking (With Tools)49.702 / 3Thinking (With Tools)+14.20
OSWorld-Verified835 / 26Thinking (With Tools)78.4010 / 26Thinking High (With Tools)+4.60
Terminal-Bench 2.17827 / 49Thinking (With Tools)76.2028 / 49Thinking High (With Tools)+1.80

Coding and Software Engineer

Gemini 3.6 Flash 2/2
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
DeepSWE4928 / 35Thinking (With Tools)3731 / 35Thinking Medium (With Tools)+12
SWE-Bench Pro - Public58.7016 / 60Thinking (With Tools)55.1032 / 60Thinking High (With Tools)+3.60

Long Context

Gemini 3.6 Flash 2/2
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
GDM-MRCR v2 (8-needle, 1M)541 / 3Thinking (No Tools)26.602 / 3Thinking (No Tools)+27.40
GDM-MRCR v2 (8-needle, 128K)91.802 / 8Thinking (No Tools)77.304 / 8Thinking (No Tools)+14.50

Math and Reasoning

Gemini 3.5 Flash 2/2
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
FrontierMath Tier 4 v221.9525 / 41Thinking High (No Tools)26.8322 / 41Thinking High (No Tools)-4.88
FrontierMath v258.9523 / 58Thinking High (No Tools)62.8121 / 58Thinking High (No Tools)-3.86

General Evaluation

Gemini 3.6 Flash 1/1
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
GPQA Diamond94.138 / 271Thinking High (No Tools)92.8023 / 271Thinking High (No Tools)+1.33

Multimodal Understanding

Gemini 3.6 Flash 1/1
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
CharXiv RQ89.404 / 19Thinking (With Tools)84.9011 / 19Thinking (With Tools)+4.50

Productivity Knowledge

Gemini 3.6 Flash 1/1
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
GDPval-AA v21,42118 / 25Thinking (No Tools)1,34920 / 25Thinking (No Tools)+72

Text Embedding

Gemini 3.6 Flash 1/1
BenchmarkGemini 3.6 FlashGemini 3.5 FlashDiff
Context Arena88.7817 / 126Thinking High (No Tools)33.49109 / 126Normal (No Tools)+55.29

Specs

FieldGemini 3.6 FlashGemini 3.5 Flash
PublisherGoogle Deep MindGoogle Deep Mind
Release date2026-07-212026-06-20
Model typeMultimodal modelMultimodal model
ArchitectureDenseDense
ParametersNot availableNot available
Context length1M1M
Max output64K64K

API pricing

Prices use DataLearner records when available; missing fields are not inferred.

ItemGemini 3.6 FlashGemini 3.5 Flash
Text input$1.5 / 1M tokens$1.5 / 1M tokens
Text output$7.5 / 1M tokens$9 / 1M tokens
Cache read$0.15 / 1M tokens$0.15 / 1M tokens

Summary

  • Gemini 3.6 Flashleads in:AI Agent - Tool Usage (3/3), Coding and Software Engineer (2/2), Long Context (2/2), General Evaluation (1/1), Multimodal Understanding (1/1), Productivity Knowledge (1/1), Text Embedding (1/1)
  • Gemini 3.5 Flashleads in:Math and Reasoning (2/2)

On average across the 13 shared benchmarks, Gemini 3.6 Flash scores 15.58 higher.

Largest single-benchmark gap: GDPval-AA v2 — Gemini 3.6 Flash 1,421 vs Gemini 3.5 Flash 1,349 (+72).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.