Summary

Rerun is an open-source platform designed to simplify the process of working with large volumes of robotics log data for training AI models, particularly in spatial and embodied AI. It offers two versions:

Rerun Open Source: This version allows users to visualize robot runs locally, create training datasets, and perform data transformations using dataframe queries and UDF support. It’s dual-licensed under MIT and Apache 2 and available on GitHub.

Rerun Data Platform: This is a managed cloud platform that handles large-scale robotics data, providing centralized data cataloging, versioning, discovery, lineage tracking, indexing, search, optimized query engines, and incremental data transformation. It’s currently in beta with selected customers.

Both versions feature data ingestion, storage, retrieval, data visualization, and data modeling capabilities. Rerun aims to reduce the time and engineering effort required to build AI training datasets by optimizing data pipelines and providing an easy-to-use interface.

Title

Rerun — Rerun

Description

Multimodal log handling and visualization for spatial and embodied AI. Managed infrastructure to ingest, store, analyze, and stream data at scale with built-in visual debugging. Fast, flexible, and easy to use.

Keywords

data, rerun, robotics, training, project, demo, scale, platform, open, source, python, sharing, infrastructure, book, visualization, time, large

Upstreams

opencv.org

NS Lookup

A 76.76.21.21

Dates

Created 2024-11-30

Updated 2026-01-27

Summarized 2026-02-28