Edgar Cervantes / Android Authority
Google’s products and services have built a reputation for being complex over time, and the company’s Gemini AI project is no different. Similar to Google Meet, which is built on top of Google Duo and Hangouts, the term Gemini can have different meanings depending on the context. For instance, many recent smartphones like the Pixel 8 and Galaxy S24 series promote having Gemini on board, but it’s not exactly the same as the Gemini chatbot accessible through an app or browser.
With various Google products and services carrying the Gemini name, it’s important to take a step back and understand what it all entails.
What is Gemini?
Calvin Wankhede / Android Authority
Gemini refers to Google’s chatbot and family of large language models (LLM) — two different but closely related aspects. The chatbot is easy to understand — it’s Google’s own ChatGPT alternative.
The Gemini chatbot can provide answers to questions, search the internet, generate images, and more. These functionalities come from the underlying AI large language model, also known as Gemini. However, Gemini is not just one single model, but rather a family of large language models available in three sizes: Nano, Pro, and Ultra.
In simpler terms, Gemini is both a chatbot and a family of large language models. Google uses the same name for both products, so understanding the context is key.
For example, when Google promotes Gemini on its Pixel smartphones, it’s referring to the language model. However, you can’t directly chat with Gemini on a Pixel just yet. Instead, the phone utilizes the AI model for features like Smart Reply. You can find a list of Gemini-powered features on the Pixel 8 if you’re interested in learning more.
Google often highlights Gemini in its chatbot form, which is how most people recognize it. The chatbot is meant to replace Google Assistant on Android, allowing users to control smart home devices, find flight deals, and more. You can also access it through a web browser for a more traditional chatbot experience.
Despite the basics, there’s more to Google’s AI project that may not be immediately apparent. From Gemini “1.5 Pro” to the Gemini Advanced subscription, here’s a breakdown of Google’s AI terminology.
Gemini Nano: The smallest language model in Google’s lineup geared towards running on smartphones and offering AI features directly on the device without requiring online processing.
Gemini Pro: A middle-sized model that powers the default chatbot experience and is comparable to OpenAI’s GPT-3.5 in terms of capabilities.
Gemini Ultra: The largest model in the family, requiring a subscription for access, and rivaling the cutting-edge GPT-4 language model.
Gemini 1.5 Pro: Google’s latest flagship AI model, offering similar capabilities to Gemini 1.0 Ultra with less computational power per word.
Gemini Advanced: An optional subscription that includes access to Gemini 1.5 Pro, along with other Google One benefits like cloud storage and Python code integration in chats.
Gemini for Workspace is an AI-powered assistant in Google apps like Gmail, Docs, Slides, and Sheets, accessible via a Gemini Advanced subscription. The Gemini chatbot can also generate AI art, supporting various languages but with limitations in specific regions and topics.
Stay informed about Google’s Gemini project by revisiting this page for updates on new features and terms!