ModelBeat
Models/Gemini 3.8 Flash
All models
G

Google DeepMind: Gemini 3.8 Flash

google/gemini-3.8-flash

Gemini 3.8 Flash is Google DeepMind’s proprietary AI model, released Sep 2, 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 95th percentile of the models Model Beat tracks.

Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.

Context
1M
Input
$1.50 / 1M
Output
$7.50 / 1M
Released
Sep 2, 2026
Modalities
textimagevideofileaudiotext
Strongest atReasoning & Knowledge top 5%Math top 10%Coding top 13%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Gemini 3.8 Flash lands among all models tracked on Model Beat.

91.0
Intelligence Index
Epoch AI
91st percentile of tracked models
87.0
Coding Index
Epoch AI
87th percentile of tracked models

Reasoning

3 evals
GPQA Diamond95.4%

Graduate-level scientific reasoning

Humanity's Last Exam47.8%

Frontier of human expert knowledge

SimpleQA Verified69.7%

Factual accuracy & hallucination

Coding

2 evals
SciCode56.6%

Scientific research coding

WebDev Arena1567

Human-rated web development

Math

1 evals
AIME 2024/202598.9%

Olympiad-qualifier math

In the news

2

Changelog

1
  • Gemini 3.8 Flash: reported SciCode score rose from 53.6% to 56.6% (+6%)benchmark · Sep 5, 2026

Compare Gemini 3.8 Flash with

Frequently asked questions

What is Gemini 3.8 Flash?

Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.

Who created Gemini 3.8 Flash?

Gemini 3.8 Flash is an AI model developed by Google DeepMind, part of the Gemini family, released Sep 2, 2026.

How much does Gemini 3.8 Flash cost?

Gemini 3.8 Flash's list price from Google DeepMind is $1.50 per million input tokens and $7.50 per million output tokens.

What is Gemini 3.8 Flash's context window?

Gemini 3.8 Flash supports a context window of up to 1M tokens.

How does Gemini 3.8 Flash perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Gemini 3.8 Flash scores 95.4% on GPQA Diamond, 47.8% on Humanity's Last Exam, 56.6% on SciCode.

Is Gemini 3.8 Flash open source?

Gemini 3.8 Flash is a proprietary model (API access), available through its provider's API rather than as a downloadable open-weight model.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Sep 10, 2026.