The Automotive Foundation Model
Delivering the most advanced computer vision technology to accelerate the development of autonomous vehicles.
Test Drive AFM-1 Now
AFM-1 Capabilities
Higher-quality, lower-cost data
Inject higher-quality pre-labels into tasks
Utilize more active tooling to produce data faster
Utilize ML quality models to ensure label accuracy
Increase development velocity
Rapidly experiment with new taxonomies
Create living datasets that can be updated as the taxonomy changes
Rapidly find the data you need
Search with open vocabulary queries for features inside or outside your taxonomy
AFM-1 Architecture
A single, open vocabulary model for segmentation and detection.
Key Features
State-of-The-Art Computer Vision
Continuous scene understanding and object detection for save AV operations.
Adaptable Architecture
Rapidly customized to new environments, camera setups, object classes.
Accelerate AV Development
10x faster labeling and simulation testing speeds time-to-market.
Reduces Costs
Automates labeling needs on new projects and reduces manual labeling costs.
Seamless Integration
Leverage Scale’s human-in-the-loop system for quality-guaranteed evaluation data.
Learn more about AFM-1
Introducing Scale’s Automotive Foundation Model
State of the art, language-grounded, perception model to accelerate the development of autonomous vehicles.
Learn More →
