Open models.
One endpoint.
Pay per token.

Princeps Sator is a pay-as-you-go inference API for open models. OpenAI-compatible, prepaid credits, live usage dashboard.

Get an API keyBrowse models →

No subscription. Top up from $10, spend by the token.

Model catalog

api.princep.org/v1

  • GPT 5.6 Luna

    1M context

    $0.17 / $1.02 per M

  • DeepSeek V4 Flashpopular

    1M context

    $0.07 / $0.14 per M

  • Kimi K3

    1M context

    $2.55 / $12.75 per M

  • GLM-5.2

    1M context

    $0.70 / $2.20 per M

20 models availableSee all →

Works with the tools you already use

Claude CodeCursorClineOpenCodeRoo CodeKilo CodeGooseCrushDroid
TRAEQoderZCodePiClaude for IDEOpenClawEigentHermes AgentSillyTavern

…and anything else that speaks OpenAI or Anthropic.

Drop-in OpenAI compatibility

Change the base URL, keep your code. Streaming and tool calls included; Anthropic-style Messages too.

client = OpenAI(base_url="https://api.princep.org/v1", api_key="sk-sator-...")

Prepaid credits

Top up from $10 and spend per token at the listed rates. Nothing recurring, nothing to cancel.

Usage dashboard

Spend by day, model and API key — the same metering your balance is billed from.

Keys under your control

Named keys, shown once, revocable anytime. Per-key usage attribution built in.

Part of Princeps

Built and run by Princeps, on the same infrastructure as the rest of the platform.

Simple, public, per-token pricing

Every model's input and output rate is on one page. What the meter says is what you pay.

View the price list