Today, we’re announcing the expansion of code assistance features to all Colab users, including users on our free-of-charge plans. Anyone in eligible locales can now try AI-powered code assistance in Colab.
Colab started as a tool built by a small team at Google Research who felt there was a better way to collaborate and iterate in the field of machine learning. Now it’s grown to more than 10 million monthly active users, including millions of students globally, making it Google's largest AI-enabled coding tool. Colab provides developers with frictionless access to powerful computing resources, free of charge, without having to install or manage any software. With its smooth Google Drive integration, it's one of the easiest ways to start programming in Python.
Democratizing machine learning globally
Colab aims to increase access to advanced computational resources for users around the world, providing access to TPUs and NVIDIA GPUs free of charge. Typically these resources are expensive and beyond the reach of many global developers and researchers, limiting their ability to pursue their projects. By providing access to compute power and offering AI-assisted coding features, Colab helps ensure all users, from advanced professionals to students, can learn, develop, and practice machine learning and AI. It’s part of our long-standing tradition to provide access to AI development tools to users and teams around the world.
We recently visited the HausaNLP team in Kano, Nigeria to learn more about how they use Colab towards the goal of democratizing access to AI. The team is focused on research that increases representation of the 300+ languages spoken in Nigeria in LLMs (large language models). Without this representation, it’s nearly impossible to build effective and inclusive AI.
Access to AI-powered code assistance
Colab supports generating code from natural language and a code-assisting chatbot. Colab is now offering a limited-time trial of these features to everyone in eligible locales to allow anyone to experience the power of Google's generative models.1 We expect this to last well into 2024, while capacity is available.
Since we added code assistance features inside Colab Pro earlier this year, developers in over 175 locales around the world have started to use this feature to increase programming speed, quality and comprehension.
One of our most used features is “Explain error” — whenever you encounter an execution error in Colab, we'll use the AI-powered chatbot to give you an explanation along with a potential fix:
We're excited to expand these code assistance features and hope it’s helpful as our users leverage Colab to learn, develop and practice their skills.