Platform
Train, evaluate, and deploy AI models with our frontier research infrastructure.Hosted Training
Run large-scale RL training on-demand, without managing infrastructure
Environments Hub
Create, share, and discover RL environments for training and evaluation
Evaluations
Run evaluations on your environments with hosted inference
Sandboxes
Secure code execution environments for AI agents
Guides
Step-by-step workflows for training and environments
Libraries
Open source tooling for environments, evaluations, and large-scale RL training.verifiers
Our library for environments and evals
prime-rl
Our large-scale async RL framework
Inference
Use hosted models through an OpenAI-compatible API.Get started
Make your first inference request
GLM-5.3
Use GLM-5.3 hosted on Prime Inference
Compute
Access GPUs and infrastructure for AI workloads.On-Demand Cloud
Deploy single GPU instances in under a minute
Storage
Persistent storage for datasets and checkpoints
Reserved Clusters
Dedicated GPU clusters with monitoring
Resources
API Reference
REST API documentation
CLI Reference
Prime CLI commands
Compute FAQ
Common questions