PRICING
Choose a path: start with InferX Free, add usage credits as needed, or deploy the InferX Platform with your infrastructure team.
Start building with hosted models on InferX through OpenAI-compatible APIs. Data processing and retention are governed by the applicable public endpoint service configuration and provider policies.
Add prepaid credits anytime
End-to-end deployment of your inference stack.
*Savings depend on workload patterns, model size, traffic shape, and existing GPU utilization.
SERVERLESS ECONOMICS
Always on. Always billing.
Paying for idle capacity.
Active compute only. Scale to zero between requests.
Pay only for active compute.
Rent GPU capacity for custom deployments and dedicated inference endpoints. Rates are measured per GPU-hour, so you pay for the capacity you use.
Hosted production models with usage measured by input and output tokens.
Dedicated model capacity priced around model and GPU configuration.
Commercial licensing for customer-operated and provider infrastructure.
Confirmed information about billing, commitments, credits, and support.
PRICING FAQ
InferX Free gives developers access to hosted models and OpenAI-compatible APIs. Data processing and retention are governed by the applicable public endpoint service configuration and provider policies.
Yes. Pay As You Go lets you add usage credits when you need them without a monthly subscription.
Enterprise Platform includes end-to-end deployment of the InferX runtime across private cloud, on-prem, bare metal, GPU cloud, or existing Kubernetes infrastructure.
Unused credits do not currently roll over.
