Google: Gemini 2.5 Pro

Google: Gemini 2.5 Pro is a text model from Google, listed on api14 under the model id google/gemini-2.5-pro. It is Google's state-of-the-art model built for advanced reasoning, coding, mathematics and scientific tasks, and it uses thinking capabilities to reason through responses with enhanced accuracy. The provider release date is 2025-06-17, and the knowledge cutoff is 2025-01-31.

The model accepts text, image, file, audio and video as input and returns text. Its context window is 1,048,576 tokens, with a maximum output of 65,536 tokens. Supported request parameters are include_reasoning, max_tokens, reasoning, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools and top_p. Because input spans several modalities while output stays text, it fits workflows where you feed rich source material and want a written answer.

That combination points to practical uses such as code generation and review, multi-step mathematical or scientific problem solving, and document analysis where long inputs matter. Since images, files, audio and video can all be supplied as input, a request can mix a PDF, a screenshot or a recording with a text prompt and receive a single text response. The large context window suits codebases, long reports and extended transcripts, and the reasoning parameter lets you control how much thinking the model applies.

You call the model through api14's OpenAI-compatible API using the model id google/gemini-2.5-pro. Requests follow the OpenAI chat completions shape, so existing OpenAI SDKs and clients work with a changed base URL and model id. Set max_tokens to cap the output, temperature and top_p to tune sampling, and stop to halt generation on given sequences. Use response_format or structured_outputs when you need structured replies, seed for reproducible sampling, and tools with tool_choice for function calling. The reasoning and include_reasoning parameters relate to the model's thinking behaviour.

Considerations follow from the facts above. Output is text only, so tasks that need generated images, audio or video are not covered here. The maximum output of 65,536 tokens is smaller than the context window, so very long single responses are bounded even though input can be far larger. If your workload is simple text generation where reasoning depth is unnecessary, a lighter text model may be a better fit.