Search⌘ K
AI Features
Home
Courses
Build AI Applications with LangChainGo

Gemini AI

Explore Google's Gemini AI and PaLM models, focusing on their features, API setup, and practical examples. Understand how to use Gemini for multimodal inputs including images, and PaLM for text-based problem solving. Gain insight into building applications with these foundation models using practical Go language demonstrations.

We'll cover the following...

Gemini AI is a large language model (LLM) developed by Google AI. It is a factual language model trained on a massive dataset of text and code. It can generate text, translate languages, write creative content, and answer queries informally.

Gemini API models

The Gemini API supports two generative AI models:

  • Gemini

  • PaLM

Gemini has an advantage over PaLM because it can also handle image input. It’s possible to provide text, or images (or both) as prompts to the Gemini models. PaLM models, on the other hand, can only handle text input and output.

Gemini models

Here are the different Gemini models, along with their features:

  1. Gemini Pro: It generates output text based on input text and can handle a multi-turn conversational format.

  2. Gemini Vision Pro: It can accept multimodal inputs (text and image) and generate text as output.

  3. AQA: It performs attributed question answering. It is trained to return answers to questions based on provided sources and estimate answerable probability.

  4. Embedding: It generates text embeddings for the input text. ...

PaLM models