@visgate_ai/client 0.2.22 → 0.3.4

package/dist/index.d.ts

@@ -175,6 +175,124 @@ declare class Models {
     search(query: string, limit?: number): Promise<ModelsResponse>;
 }
 
+/** Deployment management: /deployments endpoints */
+
+interface DeploymentCreateOptions {
+    modelName?: string;
+    hfModelId?: string;
+    gpuTier?: string;
+    region?: string;
+    hfToken?: string;
+    runpodKey?: string;
+    webhookUrl?: string;
+    cacheScope?: "off" | "shared" | "private";
+    provider?: string;
+    task?: string;
+    userS3Url?: string;
+    userAwsAccessKeyId?: string;
+    userAwsSecretAccessKey?: string;
+    userAwsEndpointUrl?: string;
+}
+interface DeploymentInfo {
+    deploymentId: string;
+    modelId?: string | null;
+    status: string;
+    endpointUrl?: string | null;
+    provider?: string | null;
+    createdAt?: string | null;
+    gpuAllocated?: string | null;
+    modelVramGb?: number | null;
+    readyAt?: string | null;
+    error?: string | null;
+}
+interface DeploymentListResponse {
+    deployments: DeploymentInfo[];
+}
+interface DeploymentGpuInfo {
+    id: string;
+    displayName: string;
+    memoryGb: number;
+    secureCloud: boolean;
+    communityCloud: boolean;
+    bidPricePerHr?: number | null;
+    pricePerHr?: number | null;
+}
+interface DeploymentGpuListResponse {
+    gpus: DeploymentGpuInfo[];
+}
+interface DeploymentLogEntry {
+    timestamp: string;
+    level: string;
+    message: string;
+}
+interface DeploymentLogsResponse {
+    deploymentId: string;
+    logs: DeploymentLogEntry[];
+}
+interface DeploymentCostResponse {
+    deploymentId: string;
+    status: string;
+    gpuAllocated?: string | null;
+    hoursRunning?: number | null;
+    pricePerHourUsd?: number | null;
+    estimatedCostUsd?: number | null;
+    note?: string | null;
+}
+interface DeploymentRunRequest {
+    input: Record<string, unknown>;
+}
+interface StreamStatusEvent {
+    deploymentId: string;
+    status?: string;
+    endpointUrl?: string;
+    estimatedRemainingSeconds?: number;
+    error?: string;
+}
+declare function deploymentInfoFromResponse(data: Record<string, unknown>): DeploymentInfo;
+declare function deploymentListResponseFromResponse(data: Record<string, unknown>): DeploymentListResponse;
+declare function deploymentGpuInfoFromResponse(data: Record<string, unknown>): DeploymentGpuInfo;
+declare function deploymentGpuListResponseFromResponse(data: Record<string, unknown>): DeploymentGpuListResponse;
+declare function deploymentLogsResponseFromResponse(data: Record<string, unknown>): DeploymentLogsResponse;
+declare function deploymentCostResponseFromResponse(data: Record<string, unknown>): DeploymentCostResponse;
+declare class Deployments {
+    private _client;
+    constructor(_client: Client);
+    create(options: DeploymentCreateOptions): Promise<DeploymentInfo>;
+    list(): Promise<DeploymentListResponse>;
+    get(deploymentId: string): Promise<DeploymentInfo>;
+    delete(deploymentId: string): Promise<void>;
+    run(deploymentId: string, input: Record<string, unknown>): Promise<Record<string, unknown>>;
+    getLogs(deploymentId: string): Promise<DeploymentLogsResponse>;
+    getCost(deploymentId: string): Promise<DeploymentCostResponse>;
+    listGpus(): Promise<DeploymentGpuListResponse>;
+    /**
+     * Streams deployment status events (SSE).
+     * Returns an abort function to stop streaming.
+     */
+    streamStatus(deploymentId: string, onEvent: (event: StreamStatusEvent) => void, onError?: (error: Error) => void): () => void;
+}
+
+/**
+ * Hugging Face Hub API — list deployable image/video generation models.
+ * Calls https://huggingface.co/api/models directly (no Visgate backend).
+ */
+interface HfModelInfo {
+    id: string;
+    modelId: string;
+    pipelineTag: string;
+    libraryName?: string;
+    likes?: number;
+    downloads?: number;
+}
+interface HfModelsListOptions {
+    task?: "text-to-image" | "image-to-image" | "text-to-video" | "image-to-video";
+    limit?: number;
+    search?: string;
+}
+declare class HfModels {
+    list(options?: HfModelsListOptions): Promise<HfModelInfo[]>;
+}
+
 /** Provider key management and balance resources */
 
 interface ProviderKeyInfo {
@@ -419,6 +537,7 @@ interface ClientOptions {
     falKey?: string | null;
     replicateKey?: string | null;
     runwayKey?: string | null;
+    runpodKey?: string | null;
 }
 interface RequestInitWithBody {
     body?: string;
@@ -437,6 +556,8 @@ declare class Client {
     readonly requests: Requests;
     readonly usage: Usage;
     readonly providers: Providers;
+    readonly deployments: Deployments;
+    readonly hfModels: HfModels;
     readonly billing: Billing;
     private _generate;
     constructor(options?: ClientOptions);
@@ -512,6 +633,6 @@ declare class VisgateConnectionError extends VisgateError {
  * Visgate JavaScript/TypeScript SDK — unified client for the Visgate vision AI gateway.
  */
 
-declare const VERSION = "0.2.2";
+declare const VERSION = "0.3.4";
 
-export { AsyncClient, type AsyncVideoAccepted, AuthenticationError, Billing, type BillingInfo, type CheckoutOptions, Client, type ClientOptions, type DashboardResponse, type FeaturedSection, Generate, type GenerateResult, type ImageResult, Images, type ImagesGenerateOptions, type ModelInfo, type ModelPricing, Models, type ModelsListOptions, type ModelsResponse, type PricingResponse, type ProviderBalanceItem, type ProviderBalancesResponse, ProviderError, type ProviderKeyInfo, type ProviderKeysResponse, type ProviderValidationResult, Providers, RateLimitError, type RequestStatusResult, Requests, Usage, type UsageSummary, VERSION, VIDEO_MODEL_PRESETS, ValidationError, type VideoResult, Videos, type VideosGenerateOptions, VisgateConnectionError, VisgateError, type VisgateErrorDetails, VisgateTimeoutError, billingInfoFromResponse, featuredSectionFromResponse, generateResultFromResponse, imageResultFromResponse, modelInfoFromResponse, modelPricingFromResponse, modelsResponseFromResponse, pricingResponseFromResponse, providerBalanceItemFromResponse, providerBalancesResponseFromResponse, providerKeyInfoFromResponse, providerKeysResponseFromResponse, providerValidationResultFromResponse, requestStatusFromResponse, usageSummaryFromResponse, videoResultFromResponse };
+export { AsyncClient, type AsyncVideoAccepted, AuthenticationError, Billing, type BillingInfo, type CheckoutOptions, Client, type ClientOptions, type DashboardResponse, type DeploymentCostResponse, type DeploymentCreateOptions, type DeploymentGpuInfo, type DeploymentGpuListResponse, type DeploymentInfo, type DeploymentListResponse, type DeploymentLogEntry, type DeploymentLogsResponse, type DeploymentRunRequest, Deployments, type FeaturedSection, Generate, type GenerateResult, type HfModelInfo, HfModels, type HfModelsListOptions, type ImageResult, Images, type ImagesGenerateOptions, type ModelInfo, type ModelPricing, Models, type ModelsListOptions, type ModelsResponse, type PricingResponse, type ProviderBalanceItem, type ProviderBalancesResponse, ProviderError, type ProviderKeyInfo, type ProviderKeysResponse, type ProviderValidationResult, Providers, RateLimitError, type RequestStatusResult, Requests, type StreamStatusEvent, Usage, type UsageSummary, VERSION, VIDEO_MODEL_PRESETS, ValidationError, type VideoResult, Videos, type VideosGenerateOptions, VisgateConnectionError, VisgateError, type VisgateErrorDetails, VisgateTimeoutError, billingInfoFromResponse, deploymentCostResponseFromResponse, deploymentGpuInfoFromResponse, deploymentGpuListResponseFromResponse, deploymentInfoFromResponse, deploymentListResponseFromResponse, deploymentLogsResponseFromResponse, featuredSectionFromResponse, generateResultFromResponse, imageResultFromResponse, modelInfoFromResponse, modelPricingFromResponse, modelsResponseFromResponse, pricingResponseFromResponse, providerBalanceItemFromResponse, providerBalancesResponseFromResponse, providerKeyInfoFromResponse, providerKeysResponseFromResponse, providerValidationResultFromResponse, requestStatusFromResponse, usageSummaryFromResponse, videoResultFromResponse };

package/dist/index.js

@@ -324,6 +324,219 @@ var Models = class {
   }
 };
 
+// src/resources/deployments.ts
+function deploymentInfoFromResponse(data) {
+  return {
+    deploymentId: data.deployment_id ?? data.deploymentId ?? "",
+    modelId: data.model_id ?? data.modelId,
+    status: data.status ?? "unknown",
+    endpointUrl: data.endpoint_url ?? data.endpointUrl,
+    provider: data.provider ?? "runpod",
+    createdAt: data.created_at ?? data.createdAt,
+    gpuAllocated: data.gpu_allocated ?? data.gpuAllocated,
+    modelVramGb: data.model_vram_gb ?? data.modelVramGb,
+    readyAt: data.ready_at ?? data.readyAt,
+    error: data.error ?? null
+  };
+}
+function deploymentListResponseFromResponse(data) {
+  const deployments = data.deployments ?? [];
+  return { deployments: deployments.map(deploymentInfoFromResponse) };
+}
+function deploymentGpuInfoFromResponse(data) {
+  return {
+    id: data.id ?? "",
+    displayName: data.display_name ?? data.displayName ?? "",
+    memoryGb: data.memory_gb ?? data.memoryGb ?? 0,
+    secureCloud: Boolean(data.secure_cloud ?? data.secureCloud),
+    communityCloud: Boolean(data.community_cloud ?? data.communityCloud),
+    bidPricePerHr: data.bid_price_per_hr ?? data.bidPricePerHr,
+    pricePerHr: data.price_per_hr ?? data.pricePerHr
+  };
+}
+function deploymentGpuListResponseFromResponse(data) {
+  const gpus = data.gpus ?? [];
+  return { gpus: gpus.map(deploymentGpuInfoFromResponse) };
+}
+function deploymentLogsResponseFromResponse(data) {
+  const rawLogs = data.logs ?? [];
+  return {
+    deploymentId: data.deployment_id ?? data.deploymentId ?? "",
+    logs: rawLogs.map((entry) => ({
+      timestamp: entry.timestamp ?? "",
+      level: entry.level ?? "INFO",
+      message: entry.message ?? ""
+    }))
+  };
+}
+function deploymentCostResponseFromResponse(data) {
+  return {
+    deploymentId: data.deployment_id ?? data.deploymentId ?? "",
+    status: data.status ?? "unknown",
+    gpuAllocated: data.gpu_allocated ?? data.gpuAllocated,
+    hoursRunning: data.hours_running ?? data.hoursRunning,
+    pricePerHourUsd: data.price_per_hour_usd ?? data.pricePerHourUsd,
+    estimatedCostUsd: data.estimated_cost_usd ?? data.estimatedCostUsd,
+    note: data.note ?? null
+  };
+}
+var Deployments = class {
+  constructor(_client) {
+    this._client = _client;
+  }
+  async create(options) {
+    const data = await this._client._request("POST", "/deployments", {
+      body: JSON.stringify({
+        model_name: options.modelName,
+        hf_model_id: options.hfModelId,
+        gpu_tier: options.gpuTier,
+        region: options.region,
+        hf_token: options.hfToken,
+        runpod_key: options.runpodKey,
+        webhook_url: options.webhookUrl,
+        cache_scope: options.cacheScope,
+        provider: options.provider,
+        task: options.task,
+        user_s3_url: options.userS3Url,
+        user_aws_access_key_id: options.userAwsAccessKeyId,
+        user_aws_secret_access_key: options.userAwsSecretAccessKey,
+        user_aws_endpoint_url: options.userAwsEndpointUrl
+      })
+    });
+    return deploymentInfoFromResponse(data);
+  }
+  async list() {
+    const data = await this._client._request("GET", "/deployments");
+    return deploymentListResponseFromResponse(data);
+  }
+  async get(deploymentId) {
+    const data = await this._client._request(
+      "GET",
+      `/deployments/${encodeURIComponent(deploymentId)}`
+    );
+    return deploymentInfoFromResponse(data);
+  }
+  async delete(deploymentId) {
+    await this._client._request("DELETE", `/deployments/${encodeURIComponent(deploymentId)}`);
+  }
+  async run(deploymentId, input) {
+    const data = await this._client._request("POST", `/deployments/${encodeURIComponent(deploymentId)}/run`, {
+      body: JSON.stringify({ input })
+    });
+    return data;
+  }
+  async getLogs(deploymentId) {
+    const data = await this._client._request(
+      "GET",
+      `/deployments/${encodeURIComponent(deploymentId)}/logs`
+    );
+    return deploymentLogsResponseFromResponse(data);
+  }
+  async getCost(deploymentId) {
+    const data = await this._client._request(
+      "GET",
+      `/deployments/${encodeURIComponent(deploymentId)}/cost`
+    );
+    return deploymentCostResponseFromResponse(data);
+  }
+  async listGpus() {
+    const data = await this._client._request("GET", "/deployments/gpus");
+    return deploymentGpuListResponseFromResponse(data);
+  }
+  /**
+   * Streams deployment status events (SSE).
+   * Returns an abort function to stop streaming.
+   */
+  streamStatus(deploymentId, onEvent, onError) {
+    const controller = new AbortController();
+    const base = this._client.baseUrl.replace(/\/$/, "");
+    const url = `${base}/deployments/${encodeURIComponent(deploymentId)}/stream`;
+    const headers = {};
+    if (this._client.apiKey) headers.Authorization = `Bearer ${this._client.apiKey}`;
+    void (async () => {
+      try {
+        const response = await fetch(url, {
+          method: "GET",
+          headers,
+          signal: controller.signal
+        });
+        if (!response.ok || !response.body) {
+          throw new Error(`SSE stream failed with status ${response.status}`);
+        }
+        const decoder = new TextDecoder();
+        const reader = response.body.getReader();
+        let buffer = "";
+        while (true) {
+          const { value, done } = await reader.read();
+          if (done) break;
+          buffer += decoder.decode(value, { stream: true });
+          const events = buffer.split("\n\n");
+          buffer = events.pop() ?? "";
+          for (const rawEvent of events) {
+            const dataLine = rawEvent.split("\n").find((line) => line.startsWith("data:"));
+            if (!dataLine) continue;
+            const payload = dataLine.slice(5).trim();
+            if (!payload) continue;
+            try {
+              const parsed = JSON.parse(payload);
+              onEvent({
+                deploymentId: parsed.deployment_id ?? parsed.deploymentId ?? deploymentId,
+                status: parsed.status,
+                endpointUrl: parsed.endpoint_url ?? parsed.endpointUrl,
+                estimatedRemainingSeconds: parsed.estimated_remaining_seconds,
+                error: parsed.error
+              });
+            } catch {
+            }
+          }
+        }
+      } catch (err) {
+        if (!controller.signal.aborted) {
+          const error = err instanceof Error ? err : new Error(String(err));
+          onError?.(error);
+        }
+      }
+    })();
+    return () => controller.abort();
+  }
+};
+
+// src/resources/hf-models.ts
+var HF_API_BASE = "https://huggingface.co/api/models";
+function normalizeModel(raw) {
+  const id = raw.id ?? raw.modelId ?? "";
+  return {
+    id,
+    modelId: raw.modelId ?? id,
+    pipelineTag: raw.pipeline_tag ?? "",
+    libraryName: raw.library_name,
+    likes: raw.likes,
+    downloads: raw.downloads
+  };
+}
+var HfModels = class {
+  async list(options = {}) {
+    const task = options.task ?? "text-to-image";
+    const limit = Math.min(Math.max(options.limit ?? 50, 1), 100);
+    const params = new URLSearchParams();
+    params.set("pipeline_tag", task);
+    params.set("limit", String(limit));
+    if (options.search && options.search.trim()) {
+      params.set("search", options.search.trim());
+    }
+    const url = `${HF_API_BASE}?${params.toString()}`;
+    const response = await fetch(url);
+    if (!response.ok) {
+      throw new Error(`Hugging Face API error: ${response.status} ${response.statusText}`);
+    }
+    const data = await response.json();
+    if (!Array.isArray(data)) {
+      return [];
+    }
+    return data.map((item) => normalizeModel(item));
+  }
+};
+
 // src/resources/providers.ts
 function providerKeyInfoFromResponse(data) {
   return {
@@ -591,7 +804,7 @@ var DEFAULT_BASE_URL = "https://visgateai.com/api/v1";
 var DEFAULT_TIMEOUT = 12e4;
 var DEFAULT_MAX_RETRIES = 2;
 var RETRYABLE_STATUS_CODES = /* @__PURE__ */ new Set([429, 500, 502, 503, 504]);
-var SDK_VERSION = "0.2.2";
+var SDK_VERSION = "0.3.3";
 function getVersion() {
   try {
     if (typeof __VERSION__ !== "undefined") return __VERSION__;
@@ -608,6 +821,7 @@ function buildHeaders(apiKey, options) {
   if (options.falKey) headers["X-Fal-Key"] = options.falKey;
   if (options.replicateKey) headers["X-Replicate-Key"] = options.replicateKey;
   if (options.runwayKey) headers["X-Runway-Key"] = options.runwayKey;
+  if (options.runpodKey) headers["X-Runpod-Key"] = options.runpodKey;
   return headers;
 }
 function backoff(attempt) {
@@ -672,7 +886,8 @@ var Client = class {
     this.headers = buildHeaders(this.apiKey, options.proxyUrl ? {} : {
       falKey: options.falKey,
       replicateKey: options.replicateKey,
-      runwayKey: options.runwayKey
+      runwayKey: options.runwayKey,
+      runpodKey: options.runpodKey
     });
     this.images = new Images(this);
     this.models = new Models(this);
@@ -680,6 +895,8 @@ var Client = class {
     this.requests = new Requests(this);
     this.usage = new Usage(this);
     this.providers = new Providers(this);
+    this.deployments = new Deployments(this);
+    this.hfModels = new HfModels();
     this.billing = new Billing(this);
     this._generate = new Generate(this);
   }
@@ -769,8 +986,8 @@ var AsyncClient = class extends Client {
 };
 
 // src/index.ts
-var VERSION = "0.2.2";
+var VERSION = "0.3.4";
 
-export { AsyncClient, AuthenticationError, Billing, Client, Generate, Images, Models, ProviderError, Providers, RateLimitError, Requests, Usage, VERSION, VIDEO_MODEL_PRESETS, ValidationError, Videos, VisgateConnectionError, VisgateError, VisgateTimeoutError, billingInfoFromResponse, featuredSectionFromResponse, generateResultFromResponse, imageResultFromResponse, modelInfoFromResponse, modelPricingFromResponse, modelsResponseFromResponse, pricingResponseFromResponse, providerBalanceItemFromResponse, providerBalancesResponseFromResponse, providerKeyInfoFromResponse, providerKeysResponseFromResponse, providerValidationResultFromResponse, requestStatusFromResponse, usageSummaryFromResponse, videoResultFromResponse };
+export { AsyncClient, AuthenticationError, Billing, Client, Deployments, Generate, HfModels, Images, Models, ProviderError, Providers, RateLimitError, Requests, Usage, VERSION, VIDEO_MODEL_PRESETS, ValidationError, Videos, VisgateConnectionError, VisgateError, VisgateTimeoutError, billingInfoFromResponse, deploymentCostResponseFromResponse, deploymentGpuInfoFromResponse, deploymentGpuListResponseFromResponse, deploymentInfoFromResponse, deploymentListResponseFromResponse, deploymentLogsResponseFromResponse, featuredSectionFromResponse, generateResultFromResponse, imageResultFromResponse, modelInfoFromResponse, modelPricingFromResponse, modelsResponseFromResponse, pricingResponseFromResponse, providerBalanceItemFromResponse, providerBalancesResponseFromResponse, providerKeyInfoFromResponse, providerKeysResponseFromResponse, providerValidationResultFromResponse, requestStatusFromResponse, usageSummaryFromResponse, videoResultFromResponse };
 //# sourceMappingURL=index.js.map
 //# sourceMappingURL=index.js.map