ModelBeat
Models/Gemini 3.6 Flash
All models
G

Google: Gemini 3.6 Flash

google/gemini-3.6-flash

Gemini 3.6 Flash is Google’s AI model, released Jul 21, 2026. It costs $1.50/M input and $7.50/M output tokens, and handles a 1M-token context window. Its strongest use case is reasoning & knowledge, where it ranks in the 87th percentile of the models Model Beat tracks.

Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development.

Context
1M
Input
$1.50 / 1M
Output
$7.50 / 1M
Released
Jul 21, 2026
Modalities
textimagevideofileaudiotext
Strongest atReasoning & Knowledge top 13%Coding top 18%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Gemini 3.6 Flash lands among all models tracked on Model Beat.

85.0
Intelligence Index
Artificial Analysis
85th percentile of tracked models
82.0
Coding Index
Artificial Analysis
82nd percentile of tracked models

Reasoning

2 evals
GPQA Diamond92.8%

Graduate-level scientific reasoning

Humanity's Last Exam38.3%

Frontier of human expert knowledge

Coding

1 evals
SciCode52.7%

Scientific research coding

In the news

1

Changelog

1
  • Gemini 3.6 Flash (Google) added to the tracker: $1.5/$7.5 per 1M, 1M context.tracked · Jul 21, 2026

Compare Gemini 3.6 Flash with

Frequently asked questions

What is Gemini 3.6 Flash?

Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development.

Who created Gemini 3.6 Flash?

Gemini 3.6 Flash is an AI model developed by Google, part of the Gemini family, released Jul 21, 2026.

How much does Gemini 3.6 Flash cost?

Gemini 3.6 Flash's list price from Google is $1.50 per million input tokens and $7.50 per million output tokens.

What is Gemini 3.6 Flash's context window?

Gemini 3.6 Flash supports a context window of up to 1M tokens.

How does Gemini 3.6 Flash perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Gemini 3.6 Flash scores 92.8% on GPQA Diamond, 38.3% on Humanity's Last Exam, 52.7% on SciCode.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Jul 21, 2026.