Google: Gemini 3 Flash Preview

Google: Gemini 3 Flash Preview is a text model from Google, offered here as a preview release dated 2025-12-17. It sits in the Flash line, which Google describes as high speed and high value, and the provider positions it as a thinking model built for agentic workflows, multi turn chat and coding assistance. It is available on api14 under the model id google/gemini-3-flash-preview.

The model accepts text, image, file, audio and video as input, and it returns text. Its context window is 1,048,576 tokens, and the maximum output is 65,536 tokens. Supported request parameters include max_tokens, temperature, top_p, stop, seed and response_format, along with structured_outputs for schema constrained replies. Reasoning is exposed through the reasoning, reasoning_effort and include_reasoning parameters, and tool use is covered by tools and tool_choice. Google's own description notes near Pro level reasoning and tool use, so the model is intended to handle multi step tasks rather than only single turn replies.

Those capabilities point to a few practical patterns. Agentic workflows fit well because the model can call tools and reason across steps while holding a large working context. Multi turn chat is another natural use, since the long context window lets you keep a lengthy conversation or a large document set in view without trimming. Coding assistance follows from the same combination of reasoning and tool calling, for example reading a repository, proposing edits and invoking tools. Because the input side is multimodal, you can send screenshots, audio or video alongside text and ask for a text answer, which suits tasks like summarizing a recording or extracting information from an image.

To call it through api14, use the OpenAI-compatible chat completions endpoint with the model id google/gemini-3-flash-preview. You can pass the parameters listed above in the request body, including temperature, top_p, max_tokens, stop and seed for sampling control, response_format or structured_outputs when you need a defined output shape, and tools with tool_choice when the model should call functions. Reasoning behaviour is controlled with reasoning, reasoning_effort and include_reasoning. The live price table on this page covers what a request costs.

Keep the preview status in mind. It is a preview release, so behaviour and availability may change, and the output is text only, which means image, audio or video generation is not part of this model. If your task needs a different output modality or a non preview model, another entry in the catalogue may fit better.