Cua Environments Catalog
Product
Cua Environments Catalog offers scoped trajectory data packages for computer-use agents.
- Currently Offered
Cua Environments Catalog is a data catalog from Cua that packages scoped trajectories for training computer-use agents across more than 4,000 cloud desktop environments. The catalog draws on a corpus of more than 400,000 tasks to supply reproducible environment-task pairs for evaluation and fine-tuning.
The offering targets teams building computer-use agents that operate graphical interfaces through screenshots and keyboard and mouse actions. It competes with multimodal and world-model datasets and AI training data products by focusing on grounded desktop trajectories rather than generic web text.
Cua Environments Catalog offers scoped trajectory data packages for computer-use agents.
Datoric's world-model and agentic training-data catalog.
Simulated applications with expert-curated artifacts and verifiers for training AI agents on long-horizon professional workflows.
Expert-curated datasets and reinforcement learning environments for training and evaluating AI models.
Digital World Models predict and simulate agent actions in digital workflows for AI training.
Isolated environments for reinforcement learning, agent tool use, and model evaluation on GPU infrastructure.
Datoric's world-model and agentic training-data catalog.
Simulated applications with expert-curated artifacts and verifiers for training AI agents on long-horizon professional workflows.
Datoric's robotics training-data catalog for physical AI teams.
Expert-curated datasets and reinforcement learning environments for training and evaluating AI models.
Digital World Models predict and simulate agent actions in digital workflows for AI training.
Applied Compute's Agent Cloud trains enterprise-specific AI agents.
Isolated environments for reinforcement learning, agent tool use, and model evaluation on GPU infrastructure.