Models Model details

Google

Google: Gemini 3.8 Flash

google/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.

Input
$0.825

/ 1M tokens

Output
$4.125

/ 1M tokens

Context window
1.05M

tokens

Limits

Context

1,048,576

tokens

Max output

65,536

6.3% of the context window

Model facts

Author

Google

Tokenizer

Gemini

Modality

Text + Image + File + Audio + Video -> Text

Created

Sep 2, 2026

Reasoning

RequiredDefault MediumHighMediumLow

Publication status

Hugging Face: not mappedLicense: not publishedParameter count: not published

Modalities

Input · TextInput · ImageInput · VideoInput · FileInput · AudioOutput · Text

Generation controls

Include reasoningMax tokensReasoningReasoning effortResponse formatSeedStopStructured outputsTemperatureTool choiceToolsTop p

Capability scores

40.2
Agentic index

Better than 71% of models compared

76.3
Coding index

Better than 90% of models compared

40.9
Intelligence index

Better than 85% of models compared

Agent tasks

Category performance

Mobile apps
Top 18%

1233 Elo

54.4% win

Godot games
Top 16%

1273 Elo

53.0% win

Full-stack apps
Top 27%

1244 Elo

48.1% win

Web apps
Top 25%

1229 Elo

46.2% win

Slide decks
Top 65%

1178 Elo

39.8% win

Android apps
Top 78%

1080 Elo

33.5% win

Creative tasks

Category performance

SVG
Top 11%

1293 Elo

54.6% win

UI components
Top 12%

1319 Elo

53.9% win

ASCII art
Top 21%

1267 Elo

51.8% win

Games
Top 16%

1307 Elo

51.0% win

Data visualization
Top 26%

1266 Elo

50.5% win

Code
Top 16%

1305 Elo

50.3% win

Websites
Top 13%

1302 Elo

50.0% win

3D
Top 20%

1296 Elo

49.9% win