@zinn-digital/sdk 0.1.1431 → 0.1.1432

package/dist/generated/schema.d.ts

@@ -23740,6 +23740,48 @@ export interface paths {
         patch?: never;
         trace?: never;
     };
+    "/v1/ai/inference/chat": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Run one chat completion on an open model, paid from Zinn® AI credits.
+         * @description Serverless AI inference for Zinn® AI Hosting plans (every AI Hosting plan grants `serverless_inference`). Runs in the request, on actual tokens, billed from the org's Zinn® AI credits: plan-included credits first spend here (this is an AI Hosting surface), then bought top-up credit. The platform's small daily included allowance never funds it, and a customer's own provider key is not used.
+         *
+         *     Gates, each before anything is sent or charged: the plan grants `serverless_inference` (else `422 FEATURE_NOT_INCLUDED`); the model is offered and priced (else `422`, `refusal.aiInference.unknownModel`); the org is in good standing (else `422 AI_NOT_IN_GOOD_STANDING`); credit is left (else `422 AI_CREDIT_REQUIRED`, `details[].can_buy`). A failed model call answers `503` with a `failure` whose class is `our_fault` and names no supplier, and charges nothing. At most 64 messages of 32,000 characters; `max_output_tokens` 1 to 4,096 (default 1,024). Requires `sites.create`.
+         */
+        post: operations["createAiInferenceChat"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/v1/ai/inference/models": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * The open models serverless AI inference offers, with the customer price.
+         * @description The models `POST /v1/ai/inference/chat` accepts (Zinn® AI Hosting, AIHOSTING slice 2), each with what one million input and output tokens CHARGE the customer in USD micros (the meter's price, so the number shown is the number the ledger debits). `default_model` is the cheapest priced model, derived from measured prices. Requires `sites.view`.
+         */
+        get: operations["listAiInferenceModels"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
     "/v1/ai/model-notices": {
         parameters: {
             query?: never;
@@ -69664,6 +69706,61 @@ export interface components {
         AiImageProviderAvailabilityList: {
             data: components["schemas"]["AiImageProviderAvailability"][];
         };
+        AiInferenceChatRequest: {
+            /** @default 1024 */
+            max_output_tokens: number;
+            messages: components["schemas"]["AiInferenceMessage"][];
+            /** @description A model from `GET /v1/ai/inference/models`; empty means the default. */
+            model?: string;
+            org_id?: components["schemas"]["Uuid"];
+        };
+        AiInferenceMessage: {
+            content: string;
+            /** @enum {string} */
+            role: "system" | "user" | "assistant";
+        };
+        AiInferenceModel: {
+            /** @description The model id to send as `model`. */
+            id: string;
+            /**
+             * Format: int64
+             * @description What one million input tokens charge the customer, in USD micros.
+             */
+            input_micros_per_mtok: number;
+            /**
+             * Format: int64
+             * @description What one million output tokens charge the customer, in USD micros.
+             */
+            output_micros_per_mtok: number;
+        };
+        AiInferenceModels: {
+            /** @description The cheapest priced model. */
+            default_model: string;
+            models: components["schemas"]["AiInferenceModel"][];
+        };
+        AiInferenceReply: {
+            /**
+             * Format: int64
+             * @description The org's Zinn® AI credit balance after the call, in USD micros.
+             */
+            balance_micros: number;
+            /**
+             * Format: int64
+             * @description What this call charged, in USD micros.
+             */
+            charged_micros: number;
+            finish_reason: string;
+            message: {
+                content: string;
+                /** @enum {string} */
+                role: "assistant";
+            };
+            model: string;
+            usage: {
+                input_tokens: number;
+                output_tokens: number;
+            };
+        };
         /** @description One model, as read from the provider's own catalogue and proven against our prompts. ⛔ Model ids are **data, never a hardcoded list** (§2.2) — this table is refreshed daily from each vendor. */
         AiModel: {
             /** @description 0 means the provider did not tell us. */
@@ -151243,6 +151340,62 @@ export interface operations {
             429: components["responses"]["TooManyRequests"];
         };
     };
+    createAiInferenceChat: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["AiInferenceChatRequest"];
+            };
+        };
+        responses: {
+            /** @description The reply, the tokens used, what was charged and the balance after. */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AiInferenceReply"];
+                };
+            };
+            401: components["responses"]["Unauthorized"];
+            403: components["responses"]["Forbidden"];
+            422: components["responses"]["UnprocessableEntity"];
+            429: components["responses"]["TooManyRequests"];
+            503: components["responses"]["ServiceUnavailable"];
+        };
+    };
+    listAiInferenceModels: {
+        parameters: {
+            query?: {
+                /** @description Which organization this is about. Required when the caller belongs to more than one; a caller in exactly one may omit it. */
+                org_id?: components["schemas"]["Uuid"];
+            };
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description The offered models and their prices. */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AiInferenceModels"];
+                };
+            };
+            401: components["responses"]["Unauthorized"];
+            403: components["responses"]["Forbidden"];
+            422: components["responses"]["UnprocessableEntity"];
+            429: components["responses"]["TooManyRequests"];
+        };
+    };
     listAiModelNotices: {
         parameters: {
             query?: never;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@zinn-digital/sdk",
-  "version": "0.1.1431",
+  "version": "0.1.1432",
   "description": "Typed TypeScript client for the Zinn Digital Platform API, generated from its OpenAPI spec.",
   "license": "MIT",
   "author": "Neil Lock \u2014 CEO, Zinn Digital\u00ae Ltd <office@zinndigital.com>",