AI models for Gemini API

The Gemini API provides a programming interface for both the Gemini and PaLM generative AI models. Gemini is Google's latest generation of generative models, and goes beyond the capabilities of the PaLM family of models. This page provides general guidance on how to choose a model for your application. This guide provides information about each model family and variations to help you decide which is the best fit for your use case. Here is a quick summary of the available models and their capabilities:

Models Input Output
  • Gemini Pro
Text Text
  • Gemini Pro Vision
Text and images Text
PaLM (legacy)
  • Bison Text
Text Text
  • Bison Chat
Text Text
  • Embedding
Text Text embeddings
  • AQA
Text Text

A key difference between the Gemini and PaLM models is that the Gemini vision model is able to handle image input. You can prompt Gemini models with text, or images, or both. PaLM models only handle text input and output. Both families of models can perform text prompting, chat interactions, and structured prompting.

Important: The current release of Gemini models do not support the following features, which are supported by the PaLM models:

  • Tuning the language model with your own data to perform specific tasks

Safety and intended use

Generative artificial intelligence models are powerful tools, but they are not without their limitations. Their versatility and applicability can sometimes lead to unexpected outputs, such as outputs that are inaccurate, biased, or offensive. Post-processing, and rigorous manual evaluation are essential to limit the risk of harm from such outputs. See the safety guidance for additional safe use suggestions.

The models provide by the Gemini API can be used for a wide variety of generative AI and natural language processing (NLP) applications. Use of these functions is only available through the Gemini API or the Google AI Studio web app. Your use of Gemini API is also subject to the Generative AI Prohibited Use Policy and the Gemini API terms of service.