Google introduces Gemma open source AI models


  • After the recent OpenAI’s text-to-video model Sora dominated news feeds last week, Google has now introduced Gemma which is its latest open artificial intelligence (AI) offering.
  • Google has, in the past few months, introduced its Gemini models which are large and mid-sized models meant for complex tasks.
  • The newly unveiled Gemma, however, is a lightweight, smaller model and is aimed at helping developers worldwide build AI responsibly, in compliance with Google’s AI principles.

What is Gemma?

  • Gemma is a family of lightweight state-of-the-art open models which has been built using the same research and technology used in Gemini models by Google DeepMind, and other teams across Google.
  • Google has informed that that its name is derived from the Latin word ‘gemma,’ which translates to precious stone.
  • Gemma is being offered currently in two model sizes, Gemma 2B and Gemma 7B, which have been released with pre-trained and instruction-tuned variants.
  • Along with Gemma, Google has also released a new Responsible Generative AI toolkit which provides guidance and essential tools for creating safer AI applications with Gemma.
  • Google has informed that the pre-trained and instruction-tuned Gemma models can run on laptops, workstations, or even on Google Cloud with easy deployment on Vertex AI and Google Kubernetes Engine (GKE).
  • Gemma’s optimisation for multiple AI hardware platforms ensures industry-leading performance which include NVIDIA GPUs and Google Cloud TPUs.

How does Gemma perform?

  • Google has noted that Gemma shares some key technical and infrastructure components with Gemini, which is its most capable AI model to date.
  • Because of its underlying technology, both Gemma 2B and Gemma 7B are capable of achieving the ‘best-in-class performance’ for their sizes when compared to the other open models.
  • Gemma has reportedly outperformed significantly larger models on key benchmarks while complying with rigorous standards for safe and also responsible outputs.

What is meant by Gemma is responsible by design’?

  • Google claims that Gemma has been designed in compliance with its various AI principles.
  • The tech giant Google has said that to make Gemma pre-trained models safe and reliable, they used automated techniques to filter certain personal information and sensitive data from its training sets.
  • Google has also fine-tuned Gemma’s models with human feedback to promote responsible behaviours, and also conducted thorough evaluations, including manual and automated testing to minimise risk.
  • Besides this, Google is also providing a toolkit alongside Gemma to help developers prioritise safety in the AI applications.
  • This toolkit comes with methods for building safety classifiers, debugging tools, and also guidance based on Google’s experience in developing large language models.

The post Google introduces Gemma open source AI models appeared first on Vajirao IAS.


Source link

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *