Rerun

Rerun is the data stack for Physical AI (spatial and embodied AI), offering open-source log handling and visualization, along with managed infrastructure to ingest, store, analyze, and stream data at scale with built-in visual debugging.

Main Features & Workflow:

1. Model: Use the SDK to model data and write it to storage or a live viewer. Rerun's data model, a time-aware Entity Component System, makes common scenarios simple but is flexible enough to handle custom data. Supports C++, Python, and Rust.
2. View: Get powerful interactive visualizations of live and recorded data. Time travel through data with the industry's fastest multimodal viewer. It's open source and runs both natively and in the browser.
3. Build: Build layouts and customize visualizations directly through code or interactively in the UI. Build visualization into tools by embedding the viewer in Rust, JS, or Gradio apps.
4. Query: Use Rerun's query APIs to extract time-aligned datasets from messy robotics recordings. Returns Apache Arrow data that plugs into favorite dataframe libraries for further analysis, and can be written to parquet files for later training.

Core Advantages:

• Start visualizing in seconds with shockingly easy setup: simple installation with no sign-up and minimal code.
• Infrastructure to simplify the data engine: Run & record data, use the viewer to find issues, and visualize training/eval to improve and deploy.
• Used by prominent projects like Huggingface’s LeRobot, Meta's Project Aria, Deepmind's Brush, and kornia-rs as their integrated visualization tool.

Pricing & Editions:

• Open Source: Visualization and simple log handling, dual licensed under MIT and Apache 2. Includes the fastest & easiest to use multimodal visualizer (runs native, on the web, and is embeddable in notebooks and web apps) and an SDK for modeling and handling multimodal logs.
• Commercial: Data management at scale, currently under development with select design partners (requires signing up for the waitlist). Includes a Data platform for Physical AI (instantly visualizable and debuggable data), ingestion/storage engine and index management for large scale data, and dataset management for recordings and structured tables.