Julia Programming Language Now Supported in Google Colab: Unlocking New AI and Data Science Capabilities

The Julia programming language has officially gained support in Google Colab, as announced in this [link](https://juliahub.com/blog/julia-now-available-on-google-colab). It is now available as an option within the execution environments.

Previously, running Julia on Colab required workaround methods, like manually setting it up in every session. Now, users can take advantage of free T4 GPUs (compatible with CUDA.jl) to execute Julia workloads for high-performance computing, data science, and AI research. Transforming functions for GPU execution is similar to using CUDA; for instance, you would use gpu_data_array = cu(cpu_data_array).

Packages such as CUDA, Reactant, and DiffEqGPU now function seamlessly in the Julia runtime, enabling the implementation of advanced machine learning applications and scientific computing. GPU computing can significantly accelerate parallel processing tasks, such as rendering images of the Mandelbrot set.

To ensure a smooth Julia runtime experience, the Colab environment is built on version 1.10 LTS and comes pre-installed with essential packages, including IJulia (to ensure compatibility with Jupyter/Colab), CSV, and DataFrames (for data manipulation), as well as Makie and Plots (for visualization).

Moreover, developers are exploring the possibility of pre-installing CUDA to facilitate GPU acceleration in Colab.

Julia is now integrated with AI Gemini in Colab, allowing for easy incorporation of AI-based concepts into workflows.

To get started with Julia, simply open Google Colab, select «Runtime» > «Change runtime type,» and choose «Julia» as your execution environment.

Earlier, Google [released](https://habr.com/ru/news/888032/) an update for Colab that transforms how data analysis is conducted. The free cloud-based Jupyter Notebook environment now includes a Data Science Agent based on Gemini, automating tasks such as library imports, data uploads, and template code generation.