Agnes AI provider for pi — registers two providers sharing the same text + image models but differing in billing model.
| Provider ID | Name | Billing | API Key Env |
|---|---|---|---|
agnes |
Agnes AI | Token billing | $AGNES_API_KEY |
agnes-plan |
Agnes AI (Token Plan) | Subscription plan | $AGNES_PLAN_API_KEY |
Token-based billing provider. Agnes has not published official token pricing, so cost values are 0 (no cost shown). Update when Agnes publishes pricing — use the non-discounted regular-period price (see PROVIDER.md).
Subscription plan provider. Cost set to zero — the subscription fee is a fixed monthly charge, not per-token.
Both providers export the same models:
| Model | Reasoning | Input | Context | Max Output |
|---|---|---|---|---|
agnes-2.0-flash |
Yes | text, image | 256K | 64K |
agnes-1.5-flash |
No | text, image | 256K | 64K |
pi install npm:@d3ara1n/pi-provider-agnesOr add to ~/.pi/agent/settings.json:
Set your API key(s) via environment variable:
export AGNES_API_KEY="sk-..." # for agnes provider
export AGNES_PLAN_API_KEY="sk-..." # for agnes-plan providerOr configure via /login or auth.json.
None — this is a standalone provider with no pi-extension dependencies. It uses pi's built-in openai-completions streaming.
Not yet implemented. Agnes AI does not currently expose a public quota or balance API. When one becomes available, quota reporting will be added via @d3ara1n/pi-usage-block-core. See plans/pi-provider-agnes.md for the integration plan.
{ "extensions": [ "/absolute/path/to/pi-extensions/packages/pi-provider-agnes" ] }