You only pay for what you use — just actual compute time.
Collaborate on machine learning
Forever
Host unlimited public/private models and datasets
Create unlimited teams and collaborators
Notebook for your ML
Community support
Advanced collaboration for individuals and organizations
Starting at
SSO and SAML support
Notebook custom kernels with GPU support
Real-time collaborations with comments feature
Premium integrations (Snowflake, SQL Server, BigQuery, Redshift, and more)
Web-based support
For teams with more security, compliance, and flexible deployments
Deploy Inference on your own Infrastructure
Multi-region storage
Audit Log API
Advanced auditing
Managed billing with yearly commits
Deploy models on managed infrastructure
Starting at
Keep your cost low
Pay as you go for $1 or less per million tokens
Easy upgrade to dedicated endpoints
Starting
Choose your hardware type, including A10G, A100, and H100
Deploy custom LLM, embedding models, and fine-tuned models
Dedicated endpoints with network isolation
Fully-managed autoscaling
Create powerful AI models
Starting at
Automatic models training
Easy drag and drop interface
Monitoring and Evaluation
11 tasks available for Vision, NLP, and more
Models instantly available on the hub
Resources