@leaflow/sdk 0.140.0 → 0.141.0

package/README.md

@@ -18,6 +18,12 @@ npm run generate
 npm run build
 ```
 
-`orval.config.mjs` is plain official Orval configuration that recursively discovers OpenAPI entry documents. No custom generator,
-output postprocessing or compatibility forwarding layer is used. For local generation, CONTRACTS_DIR
+`orval.config.mjs` is plain official Orval configuration that recursively discovers OpenAPI entry documents. The Canopy hook chooses JSON or Blob for the chat completion response according to Content-Type,
+so JSON and SSE retain their respective response formats. No compatibility forwarding layer is used. For local generation, CONTRACTS_DIR
 may point to an existing clean checkout of the exact CONTRACTS_REF commit. CI checks out that commit.
+
+## Canopy migration
+
+Project inference and management are exported from `canopy` and `@leaflow/sdk/canopy/v1`.
+Replace `canopyInference` and `@leaflow/sdk/canopy/inference/v1` imports with the shared Canopy module.
+The OpenAI-compatible `/v1` paths, management `/api/v1` paths and scoped-token authentication are unchanged.

package/dist/canopy/v1/generated/functions.d.ts

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *
@@ -28,13 +30,99 @@
  * - **The service is enabled before use.** A project enables the model platform once, through `create-service`; the forwarding endpoints refuse its requests until the service is `active`. Requests are then billed under the service's subscription by the tokens they use, at the rates stated in each model's `pricing`.
  * OpenAPI spec version: 1.0.0
  */
-import type { APIKeyResource, APIKeyUsageListResponseBody, ConflictResponse, CreateAPIKeyRequestBody, CreateServiceRequestBody, CreateServiceResponseBody, CursorPageRequestResource, Error, GetUsageSummaryParams, GetUsageTimelineParams, IssuedAPIKeyResource, LengthAwarePageAPIKeyResource, ListApiKeysParams, ListRequestsParams, ListUsageByApiKeyParams, ListUsageByModelParams, ModelListResponseBody, ModelResource, ModelUsageListResponseBody, Quote, RequestResource, ServiceResource, TotalsResource, UpdateAPIKeyRequestBody, UsageTimelineResponseBody } from "./models.js";
+import type { APIKeyResource, APIKeyUsageListResponseBody, ChatCompletion, ChatCompletionRequest, ConflictResponse, CreateAPIKeyRequestBody, CreateServiceRequestBody, CreateServiceResponseBody, CursorPageRequestResource, EmbeddingRequest, EmbeddingResponse, Error, GetUsageSummaryParams, GetUsageTimelineParams, InferenceError, IssuedAPIKeyResource, LengthAwarePageAPIKeyResource, ListApiKeysParams, ListRequestsParams, ListUsageByApiKeyParams, ListUsageByModelParams, Model, ModelList, ModelListResponseBody, ModelResource, ModelUsageListResponseBody, Quote, RequestResource, ServiceResource, TotalsResource, UpdateAPIKeyRequestBody, UsageTimelineResponseBody } from "./models.js";
 export type HTTPStatusCode1xx = 100 | 101 | 102 | 103;
 export type HTTPStatusCode2xx = 200 | 201 | 202 | 203 | 204 | 205 | 206 | 207;
 export type HTTPStatusCode3xx = 300 | 301 | 302 | 303 | 304 | 305 | 307 | 308;
 export type HTTPStatusCode4xx = 400 | 401 | 402 | 403 | 404 | 405 | 406 | 407 | 408 | 409 | 410 | 411 | 412 | 413 | 414 | 415 | 416 | 417 | 418 | 419 | 420 | 421 | 422 | 423 | 424 | 426 | 428 | 429 | 431 | 451;
 export type HTTPStatusCode5xx = 500 | 501 | 502 | 503 | 504 | 505 | 507 | 511;
 export type HTTPStatusCodes = HTTPStatusCode1xx | HTTPStatusCode2xx | HTTPStatusCode3xx | HTTPStatusCode4xx | HTTPStatusCode5xx;
+export type createChatCompletionResponse200ApplicationJson = {
+    data: ChatCompletion;
+    status: 200;
+};
+export type createChatCompletionResponse200TextEventStream = {
+    data: Blob;
+    status: 200;
+};
+export type createChatCompletionResponseDefault = {
+    data: InferenceError;
+    status: Exclude<HTTPStatusCodes, 200>;
+};
+export type createChatCompletionResponseSuccess = (createChatCompletionResponse200ApplicationJson | createChatCompletionResponse200TextEventStream) & {
+    headers: Headers;
+};
+export type createChatCompletionResponseError = createChatCompletionResponseDefault & {
+    headers: Headers;
+};
+export type createChatCompletionResponse = createChatCompletionResponseSuccess | createChatCompletionResponseError;
+export declare const getCreateChatCompletionUrl: () => string;
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create a chat completion
+ */
+export declare const createChatCompletion: (chatCompletionRequest: ChatCompletionRequest, options?: RequestInit, fetchFn?: typeof globalThis.fetch) => Promise<createChatCompletionResponse>;
+export type createEmbeddingsResponse200 = {
+    data: EmbeddingResponse;
+    status: 200;
+};
+export type createEmbeddingsResponseDefault = {
+    data: InferenceError;
+    status: Exclude<HTTPStatusCodes, 200>;
+};
+export type createEmbeddingsResponseSuccess = createEmbeddingsResponse200 & {
+    headers: Headers;
+};
+export type createEmbeddingsResponseError = createEmbeddingsResponseDefault & {
+    headers: Headers;
+};
+export type createEmbeddingsResponse = createEmbeddingsResponseSuccess | createEmbeddingsResponseError;
+export declare const getCreateEmbeddingsUrl: () => string;
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create embeddings
+ */
+export declare const createEmbeddings: (embeddingRequest: EmbeddingRequest, options?: RequestInit, fetchFn?: typeof globalThis.fetch) => Promise<createEmbeddingsResponse>;
+export type listInferenceModelsResponse200 = {
+    data: ModelList;
+    status: 200;
+};
+export type listInferenceModelsResponseDefault = {
+    data: InferenceError;
+    status: Exclude<HTTPStatusCodes, 200>;
+};
+export type listInferenceModelsResponseSuccess = listInferenceModelsResponse200 & {
+    headers: Headers;
+};
+export type listInferenceModelsResponseError = listInferenceModelsResponseDefault & {
+    headers: Headers;
+};
+export type listInferenceModelsResponse = listInferenceModelsResponseSuccess | listInferenceModelsResponseError;
+export declare const getListInferenceModelsUrl: () => string;
+/**
+ * @summary List callable models with capabilities and parameter limits
+ */
+export declare const listInferenceModels: (options?: RequestInit, fetchFn?: typeof globalThis.fetch) => Promise<listInferenceModelsResponse>;
+export type getInferenceModelResponse200 = {
+    data: Model;
+    status: 200;
+};
+export type getInferenceModelResponseDefault = {
+    data: InferenceError;
+    status: Exclude<HTTPStatusCodes, 200>;
+};
+export type getInferenceModelResponseSuccess = getInferenceModelResponse200 & {
+    headers: Headers;
+};
+export type getInferenceModelResponseError = getInferenceModelResponseDefault & {
+    headers: Headers;
+};
+export type getInferenceModelResponse = getInferenceModelResponseSuccess | getInferenceModelResponseError;
+export declare const getGetInferenceModelUrl: (model: string) => string;
+/**
+ * @summary Get a callable model with capabilities and parameter limits
+ */
+export declare const getInferenceModel: (model: string, options?: RequestInit, fetchFn?: typeof globalThis.fetch) => Promise<getInferenceModelResponse>;
 export type listApiKeysResponse200 = {
     data: LengthAwarePageAPIKeyResource;
     status: 200;

package/dist/canopy/v1/generated/functions.js

@@ -1,3 +1,103 @@
+export const getCreateChatCompletionUrl = () => {
+    return `https://canopy.leaflow.cloud/v1/chat/completions`;
+};
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create a chat completion
+ */
+export const createChatCompletion = async (chatCompletionRequest, options, fetchFn) => {
+    const getHeaders = (h) => {
+        if (!h)
+            return {};
+        if (h instanceof Headers)
+            return Object.fromEntries(h.entries());
+        if (Symbol.iterator in h) {
+            return Object.fromEntries(Array.from(h, (entry) => Array.from(entry)));
+        }
+        const headers = {};
+        for (const [name, value] of Object.entries(h)) {
+            if (value !== undefined)
+                headers[name] = value;
+        }
+        return headers;
+    };
+    const res = await (fetchFn ?? fetch)(getCreateChatCompletionUrl(), {
+        ...options,
+        method: "POST",
+        headers: { "Content-Type": "application/json", ...getHeaders(options?.headers) },
+        body: JSON.stringify(chatCompletionRequest),
+    });
+    const body = [204, 205, 304].includes(res.status)
+        ? null
+        : res.headers.get("content-type")?.includes("application/json")
+            ? JSON.parse(await res.text())
+            : await res.blob();
+    const data = body;
+    return { data, status: res.status, headers: res.headers };
+};
+export const getCreateEmbeddingsUrl = () => {
+    return `https://canopy.leaflow.cloud/v1/embeddings`;
+};
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create embeddings
+ */
+export const createEmbeddings = async (embeddingRequest, options, fetchFn) => {
+    const getHeaders = (h) => {
+        if (!h)
+            return {};
+        if (h instanceof Headers)
+            return Object.fromEntries(h.entries());
+        if (Symbol.iterator in h) {
+            return Object.fromEntries(Array.from(h, (entry) => Array.from(entry)));
+        }
+        const headers = {};
+        for (const [name, value] of Object.entries(h)) {
+            if (value !== undefined)
+                headers[name] = value;
+        }
+        return headers;
+    };
+    const res = await (fetchFn ?? fetch)(getCreateEmbeddingsUrl(), {
+        ...options,
+        method: "POST",
+        headers: { "Content-Type": "application/json", ...getHeaders(options?.headers) },
+        body: JSON.stringify(embeddingRequest),
+    });
+    const body = [204, 205, 304].includes(res.status) ? null : await res.text();
+    const data = body ? JSON.parse(body) : {};
+    return { data, status: res.status, headers: res.headers };
+};
+export const getListInferenceModelsUrl = () => {
+    return `https://canopy.leaflow.cloud/v1/models`;
+};
+/**
+ * @summary List callable models with capabilities and parameter limits
+ */
+export const listInferenceModels = async (options, fetchFn) => {
+    const res = await (fetchFn ?? fetch)(getListInferenceModelsUrl(), {
+        ...options,
+        method: "GET",
+    });
+    const body = [204, 205, 304].includes(res.status) ? null : await res.text();
+    const data = body ? JSON.parse(body) : {};
+    return { data, status: res.status, headers: res.headers };
+};
+export const getGetInferenceModelUrl = (model) => {
+    return `https://canopy.leaflow.cloud/v1/models/${encodeURIComponent(String(model))}`;
+};
+/**
+ * @summary Get a callable model with capabilities and parameter limits
+ */
+export const getInferenceModel = async (model, options, fetchFn) => {
+    const res = await (fetchFn ?? fetch)(getGetInferenceModelUrl(model), {
+        ...options,
+        method: "GET",
+    });
+    const body = [204, 205, 304].includes(res.status) ? null : await res.text();
+    const data = body ? JSON.parse(body) : {};
+    return { data, status: res.status, headers: res.headers };
+};
 export const getListApiKeysUrl = (params) => {
     const normalizedParams = new URLSearchParams();
     Object.entries(params || {}).forEach(([key, value]) => {

package/dist/canopy/v1/generated/models.d.ts

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *
@@ -28,6 +30,73 @@
  * - **The service is enabled before use.** A project enables the model platform once, through `create-service`; the forwarding endpoints refuse its requests until the service is `active`. Requests are then billed under the service's subscription by the tokens they use, at the rates stated in each model's `pricing`.
  * OpenAPI spec version: 1.0.0
  */
+export type ChatCompletionRequestMessagesItem = {
+    [key: string]: unknown;
+};
+export interface ChatCompletionRequest {
+    /** @minLength 1 */
+    model: string;
+    /** @minItems 1 */
+    messages: ChatCompletionRequestMessagesItem[];
+    stream?: boolean;
+    /** @minimum 1 */
+    max_tokens?: number;
+    /** @minimum 1 */
+    max_completion_tokens?: number;
+    reasoning_effort?: string;
+    [key: string]: unknown;
+}
+export interface ChatCompletion {
+    [key: string]: unknown;
+}
+export type EmbeddingRequestEncodingFormat = (typeof EmbeddingRequestEncodingFormat)[keyof typeof EmbeddingRequestEncodingFormat];
+export declare const EmbeddingRequestEncodingFormat: {
+    readonly float: "float";
+    readonly base64: "base64";
+};
+export interface EmbeddingRequest {
+    /** @minLength 1 */
+    model: string;
+    /** A nonempty string, string array, token integer array, or array of token integer arrays. */
+    input: string | unknown[];
+    encoding_format?: EmbeddingRequestEncodingFormat;
+    /** @minimum 1 */
+    dimensions?: number;
+    user?: string;
+    [key: string]: unknown;
+}
+export interface EmbeddingResponse {
+    [key: string]: unknown;
+}
+export interface Model {
+    id: string;
+    object: "model";
+    created: number;
+    owned_by: string;
+    display_name: string;
+    context_length: number;
+    max_output_tokens: number;
+    supports_vision: boolean;
+    supports_tools: boolean;
+    supports_reasoning: boolean;
+    reasoning_tiers: string[];
+}
+export interface ModelList {
+    object: "list";
+    data: Model[];
+}
+export type InferenceErrorError = {
+    message: string;
+    type: string;
+    /** @nullable */
+    param?: string | null;
+    /** @nullable */
+    code?: string | null;
+    [key: string]: unknown;
+};
+export interface InferenceError {
+    error: InferenceErrorError;
+}
 /**
  * What a given `code` carries alongside the message. The keys depend on the code,
  * and a client that does not recognise one ignores it.

package/dist/canopy/v1/generated/models.js

@@ -1,3 +1,7 @@
+export const EmbeddingRequestEncodingFormat = {
+    float: "float",
+    base64: "base64",
+};
 export const APIKeyResourceStatus = {
     active: "active",
     disabled: "disabled",

package/dist/canopy/v1/generated/schemas.zod.d.ts

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *
@@ -29,6 +31,38 @@
  * OpenAPI spec version: 1.0.0
  */
 import * as zod from "zod";
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create a chat completion
+ */
+export declare const CreateChatCompletionBody: zod.ZodObject<{
+    model: zod.ZodString;
+    messages: zod.ZodArray<zod.ZodRecord<zod.ZodString, zod.ZodUnknown>>;
+    stream: zod.ZodOptional<zod.ZodBoolean>;
+    max_tokens: zod.ZodOptional<zod.ZodInt>;
+    max_completion_tokens: zod.ZodOptional<zod.ZodInt>;
+    reasoning_effort: zod.ZodOptional<zod.ZodString>;
+}, zod.z.core.$strict>;
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create embeddings
+ */
+export declare const CreateEmbeddingsBody: zod.ZodObject<{
+    model: zod.ZodString;
+    input: zod.ZodUnion<readonly [zod.ZodString, zod.ZodArray<zod.ZodUnknown>]>;
+    encoding_format: zod.ZodOptional<zod.ZodEnum<{
+        base64: "base64";
+        float: "float";
+    }>>;
+    dimensions: zod.ZodOptional<zod.ZodInt>;
+    user: zod.ZodOptional<zod.ZodString>;
+}, zod.z.core.$strict>;
+/**
+ * @summary Get a callable model with capabilities and parameter limits
+ */
+export declare const GetInferenceModelParams: zod.ZodObject<{
+    model: zod.ZodString;
+}, zod.z.core.$strict>;
 /**
  * @summary List API keys
  */

package/dist/canopy/v1/generated/schemas.zod.js

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *
@@ -29,9 +31,27 @@
  * OpenAPI spec version: 1.0.0
  */
 import * as zod from "zod";
+import { ChatCompletionRequest } from "./validators/chatCompletionRequest.zod.js";
 import { CreateAPIKeyRequestBody } from "./validators/createAPIKeyRequestBody.zod.js";
 import { CreateServiceRequestBody } from "./validators/createServiceRequestBody.zod.js";
+import { EmbeddingRequest } from "./validators/embeddingRequest.zod.js";
 import { UpdateAPIKeyRequestBody } from "./validators/updateAPIKeyRequestBody.zod.js";
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create a chat completion
+ */
+export const CreateChatCompletionBody = ChatCompletionRequest;
+/**
+ * Authenticated project inference. Billing ownership comes only from the verified project token; no Canopy API key or project header is required.
+ * @summary Create embeddings
+ */
+export const CreateEmbeddingsBody = EmbeddingRequest;
+/**
+ * @summary Get a callable model with capabilities and parameter limits
+ */
+export const GetInferenceModelParams = zod.strictObject({
+    model: zod.string(),
+});
 /**
  * @summary List API keys
  */

package/dist/canopy/v1/generated/validators/aPIKeyResource.zod.d.ts

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *

package/dist/canopy/v1/generated/validators/aPIKeyResource.zod.js

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *

package/dist/canopy/v1/generated/validators/aPIKeyUsageListResponseBody.zod.d.ts

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *

package/dist/canopy/v1/generated/validators/aPIKeyUsageListResponseBody.zod.js

@@ -2,21 +2,23 @@
  * Generated by orval 🍺
  * Do not edit manually.
  * Leaflow Canopy API
- * Management of API keys, together with the models available and the usage recorded against them.
+ * Canopy model inference, API keys, model catalogue, service enablement and recorded usage.
  *
- * **These endpoints do not perform inference.** Requests are issued to the forwarding address, which accepts the OpenAI, Anthropic and Gemini request formats:
+ * OpenAI-compatible project inference is served on this host at `/v1/chat/completions`,
+ * `/v1/embeddings`, `/v1/models` and `/v1/models/{model}`. Use `https://canopy.leaflow.cloud/v1`
+ * as the OpenAI SDK base URL. Chat completions support `stream=true` SSE.
  *
- * ```
- * POST /v1/chat/completions                    OpenAI
- * POST /v1/messages                            Anthropic
- * POST /v1beta/models/{model}:generateContent  Gemini
- * ```
+ * Management operations use `/api/v1`. The management model catalogue includes pricing;
+ * `/v1/models` returns the OpenAI model format with Canopy capabilities and parameter limits.
  *
  * ## Authentication
  *
  * Every request carries a scoped token; see the `bearerAuth` security scheme.
  *
- * These management endpoints do not accept a Canopy API key. On this userapi host, the OpenAI-compatible `/v1/chat/completions`, `/v1/embeddings`, `/v1/models` and `/v1/models/{model}` endpoints accept the same project token; see the project inference contract. The separate API-key gateway continues to accept Canopy API keys.
+ * Management and project inference accept the same verified IAM scoped token. The token is the
+ * sole billing project authority; `OpenAI-Project` and historical identity headers cannot
+ * override it. These endpoints do not accept a Canopy API key. The separate API-key gateway
+ * accepts Canopy API keys and supports OpenAI, Anthropic and Gemini request formats.
  *
  * ## Conventions
  *