Discover Google: Gemini 3 Flash Preview

A high-speed model for coding assistance and multi-turn chat.

Input: text · image · file · audio · video Output: text Context: 1,048,576 tokens Release: 2025-12-17
Gemini 3 Flash Preview, released by Google on December 17, 2025, is engineered for high-speed, agentic workflows. With a context window of 1 million tokens, it supports diverse input modalities like text, images, audio, video, and PDFs, producing text outputs. This model excels in interactive development and collaborative coding tasks, offering configurable reasoning levels and structured output. Its capabilities include automatic context caching and efficient tool use, making it a reliable choice for users seeking effective reasoning without the latency of larger models.

Use Cases

Here are a few ways teams apply Google: Gemini 3 Flash Preview in practice—from fast drafting to multimodal understanding. Adapt these ideas to your workflow.

Enhance collaborative coding tasks

Streamline interactive development

Facilitate multi-turn conversations

Assist with complex reasoning challenges

Key Features

A quick look at the capabilities that make this model useful in real projects.

1M token context window

Supports multimodal inputs

Configurable reasoning levels

Automatic context caching

Optimized for agentic workflows

Specs

Overview
Vendor
google
Model ID
google/gemini-3-flash-preview
Release
2025-12-17
Modalities & context
Input
text · image · file · audio · video
Output
text
Context
1,048,576 tokens
Parameters & defaults

Supported parameters: include_reasoning, max_tokens, reasoning, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p

Defaults: temperature 0.2, top_p 0.95

Benchmark tests: Google: Gemini 3 Flash Preview

We ran this model against a few representative prompts to show its range. Review the outputs below and be the judge.

Text

Prompt:
Write 150 words on how AI might positively upend work, leisure and creativity
AI stands to catalyze a modern Renaissance by liberating human potential from the mundane. In the workplace, AI will handle repetitive data processing and administrative overhead, shifting the human role toward high-level strategy, empathy, and complex problem-solving. This transition promises a future where "work" is defined by creative purpose rather than rote drudgery. Leisure will be similarly transformed. As productivity surges, the traditional forty-hour week may dissolve, granting us unprecedented time to pursue personal passions. AI will act as a sophisticated concierge, curating bespoke experiences and learning paths that enrich our downtime. Most profoundly, AI democratizes creativity. By serving as a powerful co-pilot, it lowers the barrier to entry for complex arts. An aspiring storyteller can use AI to world-build, while a non-musician can compose symphonies. AI doesn’t replace the creator; it expands the canvas, allowing the human spirit to focus on vision rather than technical execution. Ultimately, AI empowers us to spend our lives dreaming, not just doing.
Run this prompt on Upend.AI

Image

Prompt:
Generate an SVG of a pelican riding a bicycle.
Benchmark image
Run this prompt on Upend.AI

Ready to try Google: Gemini 3 Flash Preview?

Chat with Gemini 3 Flash Preview
up.end
/ˌəpˈend/
verb

To “upend” means to completely disrupt, overturn, or drastically change the established order or structure of something. It implies a significant shift or alteration that can potentially have far-reaching consequences. When something is upended, it is turned upside down or transformed in a way that challenges conventional norms or expectations. The term often carries a sense of innovation, transformation, and sometimes even a hint of upheaval, indicating that the changes are not just minor adjustments but rather a fundamental reimagining of the status quo.