Abstract: New ML models are often derived from existing ones (e.g., via fine-tuning, quantization or distillation), forming an ecosystem where models are _related_ to each other (e.g., partially sharing structure and parameters). Managing such a large and evolving ecosystem of model derivatives is challenging. For instance, the overhead of storing all such models is high, and models may inherit bugs from related models, making it difficult to determine which of the derived models exhibit the bug and how to update them. In this paper, we propose a model versioning and management system called MGit that makes it easier to store, test, update, and collaborate on related models. MGit introduces a lineage graph that records the relations between models, optimizations to efficiently store model parameters, and abstractions over this lineage graph that facilitate model testing, updating and collaboration. We find that MGit works well in practice: MGit is able to reduce model storage footprint by up to 7$\times$. Additionally, in a user study with 20 ML practitioners, users complete a model updating task 3$\times$ faster on average with MGit.