Rerun is an open-source data toolchain designed for logging, querying, transforming, visualizing, and training data on a unified model—tailored for use cases in computer vision and AI research. The SDK is free and open source, available for Python, Rust, and C++, and includes an interactive viewer, dataframe and SQL queries, PyTorch integration, and more. For organizations needing scalability and managed infrastructure, Rerun Hub offers a cloud-hosted, single-tenant backend with object storage streaming, authentication, and persistent catalog features. Rerun is ideal for AI/ML teams, researchers, and enterprises needing robust data and visualization pipelines.
Visit Rerun's official website for product details and getting started.
Comprehensive API reference and integration guides for using Rerun.
Insights, tutorials, and updates on Rerun and its applications in data tools.