Google
Released August 13, 2026

Gemini 3.7 Flash

Gemini 3.7 Flash by Google: production-ready model for complex coding, agentic workflows, and multi-step execution with tunable thinking and 1M token context.

Inputs
Text
Outputs
Text

Model Overview

Capabilities, design details, and architectural traits

Gemini 3.7 Flash - Most intelligent workhorse model yet for coding and agents

Gemini 3.7 Flash is Google's most capable Flash-tier model, positioned specifically for complex coding, agentic workflows, and reliable multi-step execution. It ships as a generally available, production-ready model with an emphasis on greater token efficiency and stronger multimodal reasoning compared to its predecessor.

TraitDetail
Tunable thinking levelsSupports low, medium, and high thinking levels, with medium as the default
Context and output1M token context window and 64k max output tokens
Core positioningBuilt for complex coding, agentic workflows, and reliable multi-step execution
Updated Antigravity agentShips with an updated Antigravity agent integrated into the model release
Efficiency focusGreater token efficiency and more reliable code generation than prior Flash models

Reasoning control

The tunable thinking levels let developers choose between low, medium, and high reasoning depth per request, with medium as the default. This gives explicit control over the trade-off between response thoroughness and token cost, directly relevant to the model's focus on coding and agent workflows where multi-step reasoning quality matters.

Benchmark Performance

Independent evaluations · Artificial Analysis

45.2%
Intelligence
76.1%
Coding Index
36.6%
Agentic Index

Accuracy & Capability Details

GPQA - Graduate Science94.5%
Humanity's Last Exam47.9%
SciCode - Scientific Coding57.2%
Long Context Reasoning81.7%

Compare Models Side-by-Side

Evaluate specifications, pricing, and independent benchmark indices

Model Details
General Info
ProviderGoogleAnthropicOpenAI
Release DateAugust 13, 2026September 1, 2026September 3, 2026
Knowledge Cutoff--Apr 2026
Context & Limits
Context Window--1.1M
Pricing (per 1M tokens)
Input Pricing
$0.75
Best Input Pricing
$10$10
Output Pricing
$3.75
Best Output Pricing
$50$50
Modalities
Inputs
text
textimagefile
textimage
Outputs
text
text
text
Benchmarks (0-100)
Intelligence Index45.2
56.8
Best Intelligence Index
54.7
Coding Index76.1
81.6
Best Coding Index
76.9
Agentic Index36.6
58.2
Best Agentic Index
51.6
Gemini 3.7 Flash
Claude Fable 5.1
GPT-6 Astra

GPQA Benchmark

Graduate-level reasoning and expert Q&A evaluation.

95%
Gemini 3.7 Flash
GPQA Benchmark
Score: 95%
Gemini 3.7 Flash
94%
Claude Fable 5.1
GPQA Benchmark
Score: 94%
Claude Fable 5.1
96%
GPT-6 Astra
GPQA Benchmark
Score: 96%
GPT-6 Astra

Humanity's Last Exam

Extremely difficult logical reasoning and knowledge.

48%
Gemini 3.7 Flash
Humanity's Last Exam
Score: 48%
Gemini 3.7 Flash
59%
Claude Fable 5.1
Humanity's Last Exam
Score: 59%
Claude Fable 5.1
55%
GPT-6 Astra
Humanity's Last Exam
Score: 55%
GPT-6 Astra

Long Context Reasoning

Logical reasoning over long context windows.

82%
Gemini 3.7 Flash
Long Context Reasoning
Score: 82%
Gemini 3.7 Flash
85%
Claude Fable 5.1
Long Context Reasoning
Score: 85%
Claude Fable 5.1
81%
GPT-6 Astra
Long Context Reasoning
Score: 81%
GPT-6 Astra

Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.