Napster API billing is managed through the Azure Marketplace. There are no setup fees and no minimum commitments — you pay only for the minutes your agents are active.
Pricing tiers
| Deployment type | Price per minute | Price per hour |
|---|---|---|
| Bring your own LLM (customer-hosted Azure OpenAI) | $0.01/min | $0.60/hr |
| Napster-hosted model | ~$0.058/min | ~$3.50/hr |
What counts as a billable minute?
A minute is counted from when a session starts (the agent connects and is active) until the session closes. Idle sessions still count — a session that is open but silent continues to accrue minutes. Use disableIdleTimeout: false (the default) to allow the platform to close idle sessions automatically and avoid unnecessary charges.
Choosing a deployment type
- Customer-hosted (bring your own LLM): You provide your own Azure OpenAI deployment and credentials. AI inference runs entirely in your cloud environment. Napster does not access your LLM or its data. Lower cost per minute.
- Napster-hosted: Napster manages the model infrastructure on your behalf. No Azure account or deployment needed. Higher cost per minute but zero infrastructure overhead.
You choose the deployment type when creating an API key in the dashboard.
Viewing usage
Current usage is visible in the Napster dashboard. You can also retrieve per-session cost programmatically — each session record includes a cost field in USD, populated once the session closes.
Managing your subscription
Your subscription is managed from the Azure Portal. You can view invoices, change plans, and cancel from there. For enterprise volume pricing or custom arrangements, contact partners@napster.com.
Concurrency limits
Each organisation can run up to 5 concurrent sessions per transport (WebRTC, WebSocket, VoIP, SIP), for a total of 20. For higher limits, contact support.