Skip to main content
GRID provides a comprehensive suite of AI models for robotics — covering depth estimation, object detection, segmentation, vision-language understanding, robot control, and more. The fastest way to use these models is through GRID Cortex, our cloud-hosted inference service. Install the client, set your API key, and call any model in two lines of Python.

GRID Cortex Client

Get started in 2 lines of code — unified API for all hosted models including depth, detection, segmentation, VLMs, robot control, and kinematics.

​
Browse Models by Category

For detailed per-model documentation (local inference wrappers using the grid.model.* package), browse by category:

Depth Estimation

Models for perceiving depth from images, crucial for 3D scene understanding and obstacle avoidance.

Object Detection

Identify and locate objects within an image or video stream.

Feature Matching

Models for finding corresponding points between images, essential for tasks like SLAM and image stitching.

Navigation

AI models to enable autonomous movement and path planning.

Optical Flow

Estimate the motion of objects or the camera itself between consecutive frames.

Segmentation

Partition an image into multiple segments or regions, often to distinguish objects from the background.

Simultaneous Localization and Mapping (SLAM)

Enable a robot to build a map of an unknown environment while simultaneously keeping track of its location within that map.

Tracking

Follow the movement of specific objects over time in a video sequence.

Time to Collision (TTC)

Estimate the time remaining before a potential collision, critical for safety systems.

Vision Language Action (VLA)

Models that can understand and respond to queries about visual content using natural language.

Visual Language Models (VLM)

Integrate visual information with natural language processing for enhanced understanding and interaction.

Visual Odometry (VO)

Estimate the camera’s motion by analyzing sequential camera images.