The following page may contain information related to upcoming products, features and functionality. It is important to note that the information presented is for informational purposes only, so please do not rely on the information for purchasing or planning purposes. Just like with all projects, the items mentioned on the page are subject to change or delay, and the development, release, and timing of any products, features or functionality remain at the sole discretion of GitLab Inc.
| Stage | ModelOps |
| Maturity | minimal |
| Content Last Reviewed | 2024-03-21 |
Machine Learning Operations (MLOps) aims to automate building, experimenting, testing & evaluation, deployment and management of machine learning models in production reliably and efficiently.
See brief overview video of why MLOps is difficult.
Data Scientist and ML Engineers work together in GitLab to build, experiment, deploy, monitor and keep models secure and up-to-date. Their processes are reproducible, automated, collaborative, scalable, and monitored.
They further collaborate with other product development teams in GitLab so that there is tight coordination between models and their dependent applications. Teams stay informed on status of various components and can seamlessly coordinate making changes to production systems.
Like software development, machine learning and model development require automation and collaboration to consistently and iteratively deliver value. As machine learning becomes more prevalent, the number of individuals, roles, and changes increases and it also overlaps more frequently with other software development areas. Instead of siloed workflows, bringing ML workflows into GitLab as a single collaboration platform extends the DevOps culture to data scientists, helping organizations achieve better results.
Over the course of FY25, GitLab is ramping up a team dedicated to MLOps. We will focus on bringing the most basic and necessary MLOps capabilities to market. Namely, they are the ability to manage models, a model registry, and to deploy models to production.
Here's a video overview of what we have today.
Data Scientist are not developers. They primarily work in notebooks. We want to make it easy for Data Scientist to store the models they're working on in the model registry by offering integration with popular tools like Google Collab.
The ability to manage a model within GitLab exists today. We plan to actively mature the capabilities. We also plan to merge model experiments and model registry together.
A model repository is made more useful by adding the ability to deploy to production. With deployment, users can collaborate and complete a full basic ML workflow within GitLab.
Below are a few sample projects using GitLab CI and ML:
More reference implementation ideas can be found in the MLOps with GitLab overview.
Needs to be updated
