chore: Move Azure models to azure package (#1888)

.changeset/cold-bottles-wish.md

@@ -1,5 +0,0 @@
----
-"@llamaindex/doc": patch
----
-
-Fixed broken links in docs and validate-links script

.changeset/wild-parents-look.md

@@ -0,0 +1,6 @@
+---
+"@llamaindex/azure": patch
+"@llamaindex/openai": minor
+---
+
+Move Azure models to azure package

examples/storage/azure/azure-cosmosdb.ts

@@ -8,8 +8,9 @@ import {
   AzureCosmosDBNoSqlVectorStore,
   AzureCosmosNoSqlDocumentStore,
   AzureCosmosNoSqlIndexStore,
+  AzureOpenAI,
+  AzureOpenAIEmbedding,
 } from "@llamaindex/azure";
-import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
 import {
   Document,
   Settings,
@@ -38,17 +39,13 @@ import {
     "https://cognitiveservices.azure.com/.default",
   );
 
-  const azure = {
+  Settings.llm = new AzureOpenAI({
     azureADTokenProvider,
     deployment: process.env.AZURE_DEPLOYMENT_NAME,
-  };
-  Settings.llm = new OpenAI({ azure });
-  Settings.embedModel = new OpenAIEmbedding({
-    model: process.env.EMBEDDING_MODEL,
-    azure: {
-      ...azure,
-      deployment: process.env.EMBEDDING_MODEL,
-    },
+  });
+  Settings.embedModel = new AzureOpenAIEmbedding({
+    azureADTokenProvider,
+    deployment: process.env.EMBEDDING_MODEL,
   });
   const docStore = AzureCosmosNoSqlDocumentStore.fromAadToken();
   console.log({ docStore });

examples/storage/azure/azure-openai.ts

@@ -2,7 +2,7 @@ import {
   DefaultAzureCredential,
   getBearerTokenProvider,
 } from "@azure/identity";
-import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
+import { AzureOpenAI, AzureOpenAIEmbedding } from "@llamaindex/azure";
 import "dotenv/config";
 
 const AZURE_COGNITIVE_SERVICES_SCOPE =
@@ -15,11 +15,10 @@ const AZURE_COGNITIVE_SERVICES_SCOPE =
     AZURE_COGNITIVE_SERVICES_SCOPE,
   );
 
-  const azure = {
+  const llm = new AzureOpenAI({
     azureADTokenProvider,
     deployment: process.env.AZURE_DEPLOYMENT_NAME ?? "gpt-35-turbo",
-  };
-  const llm = new OpenAI({ azure });
+  });
   // complete api
   const response1 = await llm.complete({ prompt: "How are you?" });
   console.log(response1.text);
@@ -31,11 +30,9 @@ const AZURE_COGNITIVE_SERVICES_SCOPE =
   console.log(response2.message.content);
 
   // embeddings
-  const embedModel = new OpenAIEmbedding({
-    azure: {
-      ...azure,
-      deployment: process.env.EMBEDDING_MODEL,
-    },
+  const embedModel = new AzureOpenAIEmbedding({
+    azureADTokenProvider,
+    deployment: process.env.EMBEDDING_MODEL,
   });
 
   const texts = ["hello", "world"];

examples/storage/azure/azure/index.ts

@@ -9,11 +9,12 @@ import {
 } from "@azure/search-documents";
 import {
   AzureAISearchVectorStore,
+  AzureOpenAI,
+  AzureOpenAIEmbedding,
   type FilterableMetadataFieldKeysType,
   IndexManagement,
   MetadataIndexFieldType,
 } from "@llamaindex/azure";
-import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
 import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
 import dotenv from "dotenv";
 import {
@@ -67,17 +68,13 @@ function processResults(response: NodeWithScore[], mode: VectorStoreQueryMode) {
     "https://cognitiveservices.azure.com/.default",
   );
   // You need to deploy your own embedding model as well as your own chat completion model
-  const azure = {
+  Settings.llm = new AzureOpenAI({
     azureADTokenProvider,
     deployment: process.env.AZURE_DEPLOYMENT_NAME,
-  };
-  Settings.llm = new OpenAI({ azure });
-  Settings.embedModel = new OpenAIEmbedding({
-    model: process.env.EMBEDDING_MODEL,
-    azure: {
-      ...azure,
-      deployment: process.env.EMBEDDING_MODEL,
-    },
+  });
+  Settings.embedModel = new AzureOpenAIEmbedding({
+    azureADTokenProvider,
+    deployment: process.env.EMBEDDING_MODEL,
   });
 
   // ---------------------------------------------------------

examples/storage/azure/azure_dynamic_session.ts

@@ -4,8 +4,7 @@ import {
   DefaultAzureCredential,
   getBearerTokenProvider,
 } from "@azure/identity";
-import { AzureDynamicSessionTool } from "@llamaindex/azure";
-import { OpenAI } from "@llamaindex/openai";
+import { AzureDynamicSessionTool, AzureOpenAI } from "@llamaindex/azure";
 import { ReActAgent } from "llamaindex";
 
 async function main() {
@@ -15,14 +14,10 @@ async function main() {
     "https://cognitiveservices.azure.com/.default",
   );
 
-  const azure = {
+  // configure LLM model
+  const llm = new AzureOpenAI({
     azureADTokenProvider,
     deployment: process.env.AZURE_OPENAI_DEPLOYMENT ?? "gpt-35-turbo",
-  };
-
-  // configure LLM model
-  const llm = new OpenAI({
-    azure,
   });
 
   const azureDynamicSession = new AzureDynamicSessionTool();

examples/storage/azure/cosmosdb/loadVectorData.ts

@@ -2,10 +2,11 @@ import { CosmosClient } from "@azure/cosmos";
 import { DefaultAzureCredential } from "@azure/identity";
 import {
   AzureCosmosDBNoSQLConfig,
+  AzureOpenAI,
+  AzureOpenAIEmbedding,
   SimpleCosmosDBReader,
   SimpleCosmosDBReaderLoaderConfig,
 } from "@llamaindex/azure";
-import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
 import * as dotenv from "dotenv";
 import {
   Settings,
@@ -31,23 +32,19 @@ const vectorCollectionName =
 
 // This example uses Azure OpenAI llm and embedding models
 const llmInit = {
-  azure: {
-    apiVersion: process.env.AZURE_OPENAI_LLM_API_VERSION,
-    endpoint: process.env.AZURE_OPENAI_LLM_ENDPOINT,
-    apiKey: process.env.AZURE_OPENAI_LLM_API_KEY,
-  },
+  apiVersion: process.env.AZURE_OPENAI_LLM_API_VERSION,
+  endpoint: process.env.AZURE_OPENAI_LLM_ENDPOINT,
+  apiKey: process.env.AZURE_OPENAI_LLM_API_KEY,
 };
 
 const embedModelInit = {
-  azure: {
-    apiVersion: process.env.AZURE_OPENAI_EMBEDDING_API_VERSION,
-    endpoint: process.env.AZURE_OPENAI_EMBEDDING_ENDPOINT,
-    apiKey: process.env.AZURE_OPENAI_EMBEDDING_API_KEY,
-  },
+  apiVersion: process.env.AZURE_OPENAI_EMBEDDING_API_VERSION,
+  endpoint: process.env.AZURE_OPENAI_EMBEDDING_ENDPOINT,
+  apiKey: process.env.AZURE_OPENAI_EMBEDDING_API_KEY,
 };
 
-Settings.llm = new OpenAI(llmInit);
-Settings.embedModel = new OpenAIEmbedding(embedModelInit);
+Settings.llm = new AzureOpenAI(llmInit);
+Settings.embedModel = new AzureOpenAIEmbedding(embedModelInit);
 
 // Initialize the CosmosDB client
 async function initializeCosmosClient() {

examples/storage/azure/cosmosdb/queryVectorData.ts

@@ -1,7 +1,10 @@
 import { CosmosClient } from "@azure/cosmos";
 import { DefaultAzureCredential } from "@azure/identity";
-import { AzureCosmosDBNoSQLConfig } from "@llamaindex/azure";
-import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
+import {
+  AzureCosmosDBNoSQLConfig,
+  AzureOpenAI,
+  AzureOpenAIEmbedding,
+} from "@llamaindex/azure";
 import * as dotenv from "dotenv";
 import {
   Settings,
@@ -25,23 +28,19 @@ const containerName =
   process.env.AZURE_COSMOSDB_VECTOR_CONTAINER_NAME || "vectorContainer";
 
 const llmInit = {
-  azure: {
-    apiVersion: process.env.AZURE_OPENAI_LLM_API_VERSION,
-    endpoint: process.env.AZURE_OPENAI_LLM_ENDPOINT,
-    apiKey: process.env.AZURE_OPENAI_LLM_API_KEY,
-  },
+  apiVersion: process.env.AZURE_OPENAI_LLM_API_VERSION,
+  endpoint: process.env.AZURE_OPENAI_LLM_ENDPOINT,
+  apiKey: process.env.AZURE_OPENAI_LLM_API_KEY,
 };
 
 const embedModelInit = {
-  azure: {
-    apiVersion: process.env.AZURE_OPENAI_EMBEDDING_API_VERSION,
-    endpoint: process.env.AZURE_OPENAI_EMBEDDING_ENDPOINT,
-    apiKey: process.env.AZURE_OPENAI_EMBEDDING_API_KEY,
-  },
+  apiVersion: process.env.AZURE_OPENAI_EMBEDDING_API_VERSION,
+  endpoint: process.env.AZURE_OPENAI_EMBEDDING_ENDPOINT,
+  apiKey: process.env.AZURE_OPENAI_EMBEDDING_API_KEY,
 };
 
-Settings.llm = new OpenAI(llmInit);
-Settings.embedModel = new OpenAIEmbedding(embedModelInit);
+Settings.llm = new AzureOpenAI(llmInit);
+Settings.embedModel = new AzureOpenAIEmbedding(embedModelInit);
 
 async function initializeStores() {
   // Create a configuration object for the Azure CosmosDB NoSQL Vector Store

packages/providers/openai/src/embedding.ts

@@ -2,17 +2,9 @@ import { BaseEmbedding } from "@llamaindex/core/embeddings";
 import { getEnv } from "@llamaindex/env";
 import { Tokenizers } from "@llamaindex/env/tokenizers";
 import type {
-  AzureClientOptions,
-  AzureOpenAI as AzureOpenAILLM,
   ClientOptions as OpenAIClientOptions,
   OpenAI as OpenAILLM,
 } from "openai";
-import {
-  AzureOpenAIWithUserAgent,
-  getAzureConfigFromEnv,
-  getAzureModel,
-  shouldUseAzure,
-} from "./azure.js";
 
 export const ALL_OPENAI_EMBEDDING_MODELS = {
   "text-embedding-ada-002": {
@@ -36,10 +28,7 @@ export const ALL_OPENAI_EMBEDDING_MODELS = {
 
 type ModelKeys = keyof typeof ALL_OPENAI_EMBEDDING_MODELS;
 
-type LLMInstance = Pick<
-  AzureOpenAILLM | OpenAILLM,
-  "embeddings" | "apiKey" | "baseURL"
->;
+type LLMInstance = Pick<OpenAILLM, "embeddings" | "apiKey" | "baseURL">;
 
 export class OpenAIEmbedding extends BaseEmbedding {
   /** embeddding model. defaults to "text-embedding-ada-002" */
@@ -77,9 +66,8 @@ export class OpenAIEmbedding extends BaseEmbedding {
    * @param init - initial parameters
    */
   constructor(
-    init?: Omit<Partial<OpenAIEmbedding>, "lazySession"> & {
+    init?: Omit<Partial<OpenAIEmbedding>, "session"> & {
       session?: LLMInstance | undefined;
-      azure?: AzureClientOptions;
     },
   ) {
     super();
@@ -100,52 +88,20 @@ export class OpenAIEmbedding extends BaseEmbedding {
     if (key) {
       this.embedInfo = ALL_OPENAI_EMBEDDING_MODELS[key];
     }
-    if (init?.azure || shouldUseAzure()) {
-      const azureConfig = {
-        ...getAzureConfigFromEnv({
-          model: getAzureModel(this.model),
-        }),
-        ...init?.azure,
-      };
-      this.apiKey =
-        init?.session?.apiKey ?? azureConfig.apiKey ?? getEnv("OPENAI_API_KEY");
-      this.baseURL =
-        init?.session?.baseURL ??
-        azureConfig.baseURL ??
-        getEnv("OPENAI_BASE_URL");
-      this.lazySession = async () =>
-        import("openai").then(async ({ AzureOpenAI }) => {
-          AzureOpenAI = AzureOpenAIWithUserAgent(AzureOpenAI);
 
-          return (
-            init?.session ??
-            new AzureOpenAI({
-              maxRetries: this.maxRetries,
-              timeout: this.timeout!,
-              ...this.additionalSessionOptions,
-              ...azureConfig,
-            })
-          );
+    this.apiKey = init?.session?.apiKey ?? init?.apiKey; // Don't fallback to env here, handled in lazySession
+    this.baseURL = init?.session?.baseURL ?? init?.baseURL; // Don't fallback to env here, handled in lazySession
+    this.lazySession = async () =>
+      init?.session ??
+      import("openai").then(({ OpenAI }) => {
+        return new OpenAI({
+          apiKey: this.apiKey ?? getEnv("OPENAI_API_KEY"), // Fallback here
+          baseURL: this.baseURL ?? getEnv("OPENAI_BASE_URL"), // Fallback here
+          maxRetries: this.maxRetries,
+          timeout: this.timeout!,
+          ...this.additionalSessionOptions,
         });
-    } else {
-      this.apiKey =
-        init?.session?.apiKey ?? init?.apiKey ?? getEnv("OPENAI_API_KEY");
-      this.baseURL =
-        init?.session?.baseURL ?? init?.baseURL ?? getEnv("OPENAI_BASE_URL");
-      this.lazySession = async () =>
-        import("openai").then(({ OpenAI }) => {
-          return (
-            init?.session ??
-            new OpenAI({
-              apiKey: this.apiKey,
-              baseURL: this.baseURL,
-              maxRetries: this.maxRetries,
-              timeout: this.timeout!,
-              ...this.additionalSessionOptions,
-            })
-          );
-        });
-    }
+      });
   }
 
   /**

packages/providers/openai/src/index.ts

@@ -2,3 +2,4 @@ export * from "./agent";
 export * from "./embedding";
 export * from "./llm";
 export * from "./responses";
+export { type LLMInstance } from "./utils";

packages/providers/openai/src/llm.ts

@@ -16,7 +16,6 @@ import { extractText } from "@llamaindex/core/utils";
 import { getEnv, uint8ArrayToBase64 } from "@llamaindex/env";
 import { Tokenizers } from "@llamaindex/env/tokenizers";
 import type {
-  AzureClientOptions,
   ClientOptions as OpenAIClientOptions,
   OpenAI as OpenAILLM,
 } from "openai";
@@ -37,12 +36,6 @@ import type {
   ResponseFormatJSONObject,
   ResponseFormatJSONSchema,
 } from "openai/resources/index.js";
-import {
-  AzureOpenAIWithUserAgent,
-  getAzureConfigFromEnv,
-  getAzureModel,
-  shouldUseAzure,
-} from "./azure.js";
 import {
   ALL_AVAILABLE_OPENAI_MODELS,
   isFunctionCallingModel,
@@ -86,7 +79,6 @@ export class OpenAI extends ToolCallLLM<OpenAIAdditionalChatOptions> {
   constructor(
     init?: Omit<Partial<OpenAI>, "session"> & {
       session?: LLMInstance | undefined;
-      azure?: AzureClientOptions;
     },
   ) {
     super();
@@ -103,44 +95,20 @@ export class OpenAI extends ToolCallLLM<OpenAIAdditionalChatOptions> {
     this.timeout = init?.timeout ?? 60 * 1000; // Default is 60 seconds
     this.additionalChatOptions = init?.additionalChatOptions;
     this.additionalSessionOptions = init?.additionalSessionOptions;
-    this.apiKey =
-      init?.session?.apiKey ?? init?.apiKey ?? getEnv("OPENAI_API_KEY");
-    this.baseURL =
-      init?.session?.baseURL ?? init?.baseURL ?? getEnv("OPENAI_BASE_URL");
+    this.apiKey = init?.session?.apiKey ?? init?.apiKey;
+    this.baseURL = init?.session?.baseURL ?? init?.baseURL;
 
-    if (init?.azure || shouldUseAzure()) {
-      const azureConfig = {
-        ...getAzureConfigFromEnv({
-          model: getAzureModel(this.model),
-        }),
-        ...init?.azure,
-      };
-
-      this.lazySession = async () =>
-        init?.session ??
-        import("openai").then(({ AzureOpenAI }) => {
-          AzureOpenAI = AzureOpenAIWithUserAgent(AzureOpenAI);
-
-          return new AzureOpenAI({
-            maxRetries: this.maxRetries,
-            timeout: this.timeout!,
-            ...this.additionalSessionOptions,
-            ...azureConfig,
-          });
+    this.lazySession = async () =>
+      init?.session ??
+      import("openai").then(({ OpenAI }) => {
+        return new OpenAI({
+          apiKey: this.apiKey ?? getEnv("OPENAI_API_KEY"),
+          baseURL: this.baseURL ?? getEnv("OPENAI_BASE_URL"),
+          maxRetries: this.maxRetries,
+          timeout: this.timeout!,
+          ...this.additionalSessionOptions,
         });
-    } else {
-      this.lazySession = async () =>
-        init?.session ??
-        import("openai").then(({ OpenAI }) => {
-          return new OpenAI({
-            apiKey: this.apiKey,
-            baseURL: this.baseURL,
-            maxRetries: this.maxRetries,
-            timeout: this.timeout!,
-            ...this.additionalSessionOptions,
-          });
-        });
-    }
+      });
   }
 
   get supportToolCall() {

packages/providers/openai/src/responses.ts

@@ -17,20 +17,9 @@ import {
 import type { StoredValue } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
 import { getEnv, uint8ArrayToBase64 } from "@llamaindex/env";
-import {
-  OpenAI as OpenAILLM,
-  type AzureClientOptions,
-  type ClientOptions as OpenAIClientOptions,
-} from "openai";
 
 import { wrapEventCaller } from "@llamaindex/core/decorator";
 import { Tokenizers } from "@llamaindex/env/tokenizers";
-import {
-  AzureOpenAIWithUserAgent,
-  getAzureConfigFromEnv,
-  getAzureModel,
-  shouldUseAzure,
-} from "./azure";
 import {
   ALL_AVAILABLE_OPENAI_MODELS,
   isFunctionCallingModel,
@@ -45,6 +34,11 @@ import {
   type StreamState,
 } from "./utils";
 
+import {
+  OpenAI as OpenAILLM,
+  type ClientOptions as OpenAIClientOptions,
+} from "openai";
+
 export class OpenAIResponses extends ToolCallLLM<OpenAIResponsesChatOptions> {
   model: string;
   temperature: number;
@@ -76,7 +70,6 @@ export class OpenAIResponses extends ToolCallLLM<OpenAIResponsesChatOptions> {
   constructor(
     init?: Omit<Partial<OpenAIResponses>, "session"> & {
       session?: LLMInstance | undefined;
-      azure?: AzureClientOptions;
     },
   ) {
     super();
@@ -112,39 +105,17 @@ export class OpenAIResponses extends ToolCallLLM<OpenAIResponsesChatOptions> {
     this.previousResponseId = init?.previousResponseId ?? null;
     this.truncation = init?.truncation ?? null;
 
-    if (init?.azure || shouldUseAzure()) {
-      const azureConfig = {
-        ...getAzureConfigFromEnv({
-          model: getAzureModel(this.model),
-        }),
-        ...init?.azure,
-      };
-
-      this.lazySession = async () =>
-        init?.session ??
-        import("openai").then(({ AzureOpenAI }) => {
-          AzureOpenAI = AzureOpenAIWithUserAgent(AzureOpenAI);
-
-          return new AzureOpenAI({
-            maxRetries: this.maxRetries,
-            timeout: this.timeout!,
-            ...this.additionalSessionOptions,
-            ...azureConfig,
-          });
+    this.lazySession = async () =>
+      init?.session ??
+      import("openai").then(({ OpenAI }) => {
+        return new OpenAI({
+          apiKey: this.apiKey ?? getEnv("OPENAI_API_KEY"),
+          baseURL: this.baseURL ?? getEnv("OPENAI_BASE_URL"),
+          maxRetries: this.maxRetries,
+          timeout: this.timeout!,
+          ...this.additionalSessionOptions,
         });
-    } else {
-      this.lazySession = async () =>
-        init?.session ??
-        import("openai").then(({ OpenAI }) => {
-          return new OpenAI({
-            apiKey: this.apiKey,
-            baseURL: this.baseURL,
-            maxRetries: this.maxRetries,
-            timeout: this.timeout!,
-            ...this.additionalSessionOptions,
-          });
-        });
-    }
+      });
   }
 
   get session() {

packages/providers/openai/src/utils.ts

@@ -1,5 +1,5 @@
 import type { LLM, PartialToolCall } from "@llamaindex/core/llms";
-import { AzureOpenAI as AzureOpenAILLM, OpenAI as OpenAILLM } from "openai";
+import { OpenAI as OpenAILLM } from "openai";
 import type { ChatModel } from "openai/resources.mjs";
 import { OpenAI } from "./llm";
 
@@ -173,8 +173,8 @@ export type OpenAIAdditionalChatOptions = Omit<
 >;
 
 export type LLMInstance = Pick<
-  AzureOpenAILLM | OpenAILLM,
-  "chat" | "apiKey" | "baseURL" | "responses"
+  OpenAILLM,
+  "chat" | "apiKey" | "baseURL" | "responses" | "embeddings"
 >;
 
 export type OpenAIResponsesChatOptions = Omit<

packages/providers/openai/tsconfig.json

@@ -7,7 +7,7 @@
     "outDir": "./lib",
     "tsBuildInfoFile": "./lib/.tsbuildinfo"
   },
-  "include": ["./src", "package.json"],
+  "include": ["./src"],
   "references": [
     {
       "path": "../../core/tsconfig.json"

packages/providers/storage/azure/package.json

@@ -59,7 +59,6 @@
   "devDependencies": {
     "dotenv": "^16.4.7",
     "vitest": "^2.1.5",
-    "@llamaindex/openai": "workspace:*",
     "@llamaindex/core": "workspace:*",
     "@llamaindex/env": "workspace:*",
     "@types/node": "^22.9.0"
@@ -72,6 +71,8 @@
     "@azure/cosmos": "^4.1.1",
     "@azure/identity": "^4.4.1",
     "@azure/search-documents": "^12.1.0",
+    "@llamaindex/openai": "workspace:*",
+    "openai": "^4.90.0",
     "mongodb": "^6.7.0"
   }
 }

packages/providers/storage/azure/src/index.ts

@@ -9,3 +9,5 @@ export * from "./vectorStore/AzureQueryResultSearch.js";
 export * from "./tools/AzureDynamicSessionTool.node.js";
 
 export * from "./readers/index.js";
+
+export * from "./model/index.js";

packages/providers/storage/azure/src/model/azure.ts

@@ -1,7 +1,8 @@
 import { getEnv, process } from "@llamaindex/env";
+import type { LLMInstance } from "@llamaindex/openai";
 import type FinalRequestOptions from "openai";
-import { type AzureClientOptions } from "openai";
-import pkg from "../package.json";
+import { AzureOpenAI, type AzureClientOptions } from "openai";
+import pkg from "../../package.json";
 
 // NOTE we're not supporting the legacy models as they're not available for new deployments
 // https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/legacy-models
@@ -154,3 +155,25 @@ export function AzureOpenAIWithUserAgent<K extends Constructor>(Base: K) {
     }
   };
 }
+
+export type AzureInitSession = AzureClientOptions & {
+  session?: AzureOpenAI;
+};
+
+export const lazySession = (init?: AzureInitSession) => {
+  return async () => {
+    if (init?.session) {
+      return init?.session as unknown as LLMInstance;
+    }
+    const AzureOpenAILib = AzureOpenAIWithUserAgent(AzureOpenAI);
+
+    return new AzureOpenAILib({
+      // Use base class properties for retries, timeout, etc.
+      maxRetries: init?.maxRetries ?? 10,
+      timeout: init?.timeout ?? 60 * 1000,
+      // Apply Azure specific config
+      ...getAzureConfigFromEnv(),
+      ...init,
+    }) as unknown as LLMInstance;
+  };
+};

packages/providers/storage/azure/src/model/embedding.ts

@@ -0,0 +1,18 @@
+import { OpenAIEmbedding } from "@llamaindex/openai";
+import { lazySession, type AzureInitSession } from "./azure";
+
+export class AzureOpenAIEmbedding extends OpenAIEmbedding {
+  /**
+   * Azure OpenAI Embedding
+   * @param init - initial parameters
+   */
+  constructor(
+    init?: Omit<Partial<OpenAIEmbedding>, "session"> & AzureInitSession,
+  ) {
+    super({
+      ...init,
+      session: undefined,
+    });
+    this.lazySession = lazySession(init);
+  }
+}

packages/providers/storage/azure/src/model/index.ts

@@ -0,0 +1,3 @@
+export * from "./embedding";
+export * from "./llm";
+export * from "./responses";

packages/providers/storage/azure/src/model/llm.ts

@@ -0,0 +1,12 @@
+import { OpenAI } from "@llamaindex/openai";
+import { lazySession, type AzureInitSession } from "./azure";
+
+export class AzureOpenAI extends OpenAI {
+  constructor(init?: Omit<Partial<OpenAI>, "session"> & AzureInitSession) {
+    super({
+      ...init,
+      session: undefined,
+    });
+    this.lazySession = lazySession(init);
+  }
+}

packages/providers/storage/azure/src/model/responses.ts

@@ -0,0 +1,18 @@
+import { OpenAIResponses } from "@llamaindex/openai";
+import { lazySession, type AzureInitSession } from "./azure.js";
+
+export class AzureOpenAIResponses extends OpenAIResponses {
+  /**
+   * Azure OpenAI Responses
+   * @param init - initial parameters
+   */
+  constructor(
+    init?: Omit<Partial<OpenAIResponses>, "session"> & AzureInitSession,
+  ) {
+    super({
+      ...init,
+      session: undefined,
+    });
+    this.lazySession = lazySession(init);
+  }
+}

packages/providers/storage/azure/tsconfig.json

@@ -8,7 +8,7 @@
     "tsBuildInfoFile": "./lib/.tsbuildinfo",
     "types": ["node"]
   },
-  "include": ["./src"],
+  "include": ["./src", "package.json"],
   "references": [
     {
       "path": "../../../core/tsconfig.json"

pnpm-lock.yaml

@@ -1410,9 +1410,15 @@ importers:
       '@azure/search-documents':
         specifier: ^12.1.0
         version: 12.1.0
+      '@llamaindex/openai':
+        specifier: workspace:*
+        version: link:../../openai
       mongodb:
         specifier: ^6.7.0
         version: 6.7.0(@aws-sdk/credential-providers@3.810.0)(socks@2.8.4)
+      openai:
+        specifier: ^4.90.0
+        version: 4.94.0(ws@8.18.1(bufferutil@4.0.9))(zod@3.24.4)
     devDependencies:
       '@llamaindex/core':
         specifier: workspace:*
@@ -1420,9 +1426,6 @@ importers:
       '@llamaindex/env':
         specifier: workspace:*
         version: link:../../../env
-      '@llamaindex/openai':
-        specifier: workspace:*
-        version: link:../../openai
       '@types/node':
         specifier: ^22.9.0
         version: 22.9.0