Well its all G's: Google, Gemini, Gemma...getting confused? Maybe this will help!
Google of course is the search engine company founded way back in 1998 part of the massive Alphabet Inc.
Gemini is a leading 'closed' AI model.
Gemma is an 'open source' model.
So how does this help:
Google has an initiative 'Google DeepMind'. This was the combination of two of the world's leading AI labs: Google Brain and DeepMind. This approach has resulted in new products such as Gemini and Gemma.
You can read much more about Google DeepMind and its journey here...
Gemini
Gemini is described as an 'ecosystem' and Google's most capable AI.
Importantly the Google Gemini models are built for text, images, audio, video and code. Something called 'multimodality'
At the time of this blog the latest version of Gemini is version 1.5.
The Gemini AI models come in three sizes:
- Ultra - for complex tasks
- Pro - wide ranging tasks
- Nano - an efficient model for on-device tasks
There is a lot more about Gemini from Google here...
You can sign up and engage with chatting with Gemini from Google.
This is like chatting with other large language model services like OpenAi- Chatgpt and Claude.
Gemma
Gemma is a family of lightweight, open models.
These models are built from Gemini.
These models are 'open sourced' and although they come from the closed Gemini product they benefit from the Gemini research ongoing.
The Gemma models are supported by developer tools to help developers innovate and collaborate in the AO space.
Within the Gemma family there are various sizes of model and these can be chosen based on your needs.
Documentation for using and building AI using the Gemma models is found here...
Summary
So simply put Google owned by Alphabet delivers a closed developed generative AI model called Gemini which you can sign up for and use and then delivers open-sourced developer friendly models called Gemma for your own AI development.
If you would like us to help with your web presence and business needs please get in touch here...
