Release 0.6.9 (#1252)

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

README.md

@@ -167,11 +167,13 @@ export async function chatWithAgent(
       // ... adding your tools here
     ],
   });
-  const responseStream = await agent.chat({
-    stream: true,
-    message: question,
-    chatHistory: prevMessages,
-  });
+  const responseStream = await agent.chat(
+    {
+      message: question,
+      chatHistory: prevMessages,
+    },
+    true,
+  );
   const uiStream = createStreamableUI(<div>loading...</div>);
   responseStream
     .pipeTo(

apps/docs/CHANGELOG.md

@@ -1,5 +1,25 @@
 # docs
 
+## 0.0.78
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.0.77
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.0.76
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.0.75
 
 ### Patch Changes

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docs",
-  "version": "0.0.75",
+  "version": "0.0.78",
   "private": true,
   "scripts": {
     "docusaurus": "docusaurus",

examples/anthropic/chat_interactive.ts

@@ -18,7 +18,7 @@ import readline from "node:readline/promises";
   });
   const chatEngine = new SimpleChatEngine({
     llm,
-    chatHistory,
+    memory: chatHistory,
   });
   const rl = readline.createInterface({ input, output });
 

examples/huggingface/embedding.ts

@@ -27,12 +27,10 @@ async function main() {
 
   // Query the index
   const queryEngine = index.asQueryEngine();
-  const stream = await queryEngine.query(
-    {
-      query: "What did the author do in college?",
-    },
-    true,
-  );
+  const stream = await queryEngine.query({
+    query: "What did the author do in college?",
+    stream: true,
+  });
 
   // Output response
   for await (const chunk of stream) {

examples/huggingface/embeddingApi.ts

@@ -37,12 +37,10 @@ async function main() {
 
   // Query the index
   const queryEngine = index.asQueryEngine();
-  const stream = await queryEngine.query(
-    {
-      query: "What did the author do in college?",
-    },
-    true,
-  );
+  const stream = await queryEngine.query({
+    query: "What did the author do in college?",
+    stream: true,
+  });
 
   // Output response
   for await (const chunk of stream) {

examples/multimodal/context.ts

@@ -1,4 +1,5 @@
 // call pnpm tsx multimodal/load.ts first to init the storage
+import { extractText } from "@llamaindex/core/utils";
 import {
   ContextChatEngine,
   NodeWithScore,
@@ -25,8 +26,9 @@ Settings.callbackManager.on("retrieve-end", (event) => {
   const textNodes = nodes.filter(
     (node: NodeWithScore) => node.node.type === ObjectType.TEXT,
   );
+  const text = extractText(query);
   console.log(
-    `Retrieved ${textNodes.length} text nodes and ${imageNodes.length} image nodes for query: ${query}`,
+    `Retrieved ${textNodes.length} text nodes and ${imageNodes.length} image nodes for query: ${text}`,
   );
 });
 

examples/multimodal/rag.ts

@@ -1,3 +1,4 @@
+import { extractText } from "@llamaindex/core/utils";
 import {
   getResponseSynthesizer,
   OpenAI,
@@ -16,7 +17,8 @@ Settings.llm = new OpenAI({ model: "gpt-4-turbo", maxTokens: 512 });
 // Update callbackManager
 Settings.callbackManager.on("retrieve-end", (event) => {
   const { nodes, query } = event.detail;
-  console.log(`Retrieved ${nodes.length} nodes for query: ${query}`);
+  const text = extractText(query);
+  console.log(`Retrieved ${nodes.length} nodes for query: ${text}`);
 });
 
 async function main() {
@@ -30,12 +32,10 @@ async function main() {
     responseSynthesizer: getResponseSynthesizer("multi_modal"),
     retriever: index.asRetriever({ topK: { TEXT: 3, IMAGE: 1 } }),
   });
-  const stream = await queryEngine.query(
-    {
-      query: "Tell me more about Vincent van Gogh's famous paintings",
-    },
-    true,
-  );
+  const stream = await queryEngine.query({
+    query: "Tell me more about Vincent van Gogh's famous paintings",
+    stream: true,
+  });
   for await (const chunk of stream) {
     process.stdout.write(chunk.response);
   }

packages/autotool/CHANGELOG.md

@@ -1,5 +1,25 @@
 # @llamaindex/autotool
 
+## 3.0.9
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 3.0.8
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 3.0.7
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 3.0.6
 
 ### Patch Changes

packages/autotool/examples/01_node/CHANGELOG.md

@@ -1,5 +1,28 @@
 # @llamaindex/autotool-01-node-example
 
+## 0.0.18
+
+### Patch Changes
+
+- llamaindex@0.6.9
+- @llamaindex/autotool@3.0.9
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+  - @llamaindex/autotool@3.0.8
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+  - @llamaindex/autotool@3.0.7
+
 ## 0.0.15
 
 ### Patch Changes

packages/autotool/examples/01_node/package.json

@@ -13,5 +13,5 @@
   "scripts": {
     "start": "node --import tsx --import @llamaindex/autotool/node ./src/index.ts"
   },
-  "version": "0.0.15"
+  "version": "0.0.18"
 }

packages/autotool/examples/02_nextjs/CHANGELOG.md

@@ -1,5 +1,28 @@
 # @llamaindex/autotool-02-next-example
 
+## 0.1.62
+
+### Patch Changes
+
+- llamaindex@0.6.9
+- @llamaindex/autotool@3.0.9
+
+## 0.1.61
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+  - @llamaindex/autotool@3.0.8
+
+## 0.1.60
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+  - @llamaindex/autotool@3.0.7
+
 ## 0.1.59
 
 ### Patch Changes

packages/autotool/examples/02_nextjs/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool-02-next-example",
   "private": true,
-  "version": "0.1.59",
+  "version": "0.1.62",
   "scripts": {
     "dev": "next dev",
     "build": "next build",

packages/autotool/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool",
   "type": "module",
-  "version": "3.0.6",
+  "version": "3.0.9",
   "description": "auto transpile your JS function to LLM Agent compatible",
   "files": [
     "dist",

packages/cloud/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @llamaindex/cloud
 
+## 0.2.8
+
+### Patch Changes
+
+- ac41ed3: feat: bump cloud sdk version
+
 ## 0.2.7
 
 ### Patch Changes

packages/cloud/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloud",
-  "version": "0.2.7",
+  "version": "0.2.8",
   "type": "module",
   "license": "MIT",
   "scripts": {

packages/cloud/src/reader.ts

@@ -170,6 +170,15 @@ export class LlamaParseReader extends FileReader {
   vendorMultimodalModelName?: string | undefined;
   // The API key for the multimodal API. Can also be set as an env variable: LLAMA_CLOUD_VENDOR_MULTIMODAL_API_KEY
   vendorMultimodalApiKey?: string | undefined;
+
+  webhookUrl?: string | undefined;
+  premiumMode?: boolean | undefined;
+  takeScreenshot?: boolean | undefined;
+  disableOcr?: boolean | undefined;
+  disableReconstruction?: boolean | undefined;
+  inputS3Path?: string | undefined;
+  outputS3PathPrefix?: string | undefined;
+
   // numWorkers is implemented in SimpleDirectoryReader
   stdout?: WriteStream | undefined;
 
@@ -258,13 +267,13 @@ export class LlamaParseReader extends FileReader {
       use_vendor_multimodal_model: this.useVendorMultimodalModel,
       vendor_multimodal_model_name: this.vendorMultimodalModelName,
       vendor_multimodal_api_key: this.vendorMultimodalApiKey,
-      // fixme: does these fields need to be set?
-      webhook_url: undefined,
-      take_screenshot: undefined,
-      disable_ocr: undefined,
-      disable_reconstruction: undefined,
-      input_s3_path: undefined,
-      output_s3_path_prefix: undefined,
+      premium_mode: this.premiumMode,
+      webhook_url: this.webhookUrl,
+      take_screenshot: this.takeScreenshot,
+      disable_ocr: this.disableOcr,
+      disable_reconstruction: this.disableReconstruction,
+      input_s3_path: this.inputS3Path,
+      output_s3_path_prefix: this.outputS3PathPrefix,
     } satisfies {
       [Key in keyof Body_upload_file_api_v1_parsing_upload_post]-?:
         | Body_upload_file_api_v1_parsing_upload_post[Key]

packages/community/CHANGELOG.md

@@ -1,5 +1,13 @@
 # @llamaindex/community
 
+## 0.0.40
+
+### Patch Changes
+
+- 50e6b57: feat: add Amazon Bedrock Retriever
+- Updated dependencies [8b7fdba]
+  - @llamaindex/core@0.2.6
+
 ## 0.0.39
 
 ### Patch Changes

packages/community/README.md

@@ -7,6 +7,7 @@
 - Bedrock support for the Anthropic Claude Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
 - Bedrock support for the Meta LLama 2, 3 and 3.1 Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
 - Meta LLama3.1 405b tool call support
+- Bedrock support for querying Knowledge Base
 
 ## LICENSE
 

packages/community/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/community",
   "description": "Community package for LlamaIndexTS",
-  "version": "0.0.39",
+  "version": "0.0.40",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",
@@ -47,6 +47,7 @@
   },
   "dependencies": {
     "@aws-sdk/client-bedrock-runtime": "^3.642.0",
+    "@aws-sdk/client-bedrock-agent-runtime": "^3.642.0",
     "@llamaindex/core": "workspace:*",
     "@llamaindex/env": "workspace:*"
   }

packages/community/src/index.ts

@@ -3,3 +3,4 @@ export {
   BEDROCK_MODEL_MAX_TOKENS,
   Bedrock,
 } from "./llm/bedrock/index.js";
+export { AmazonKnowledgeBaseRetriever } from "./retrievers/bedrock.js";

packages/community/src/retrievers/bedrock.ts

@@ -0,0 +1,165 @@
+import type { KnowledgeBaseVectorSearchConfiguration } from "@aws-sdk/client-bedrock-agent-runtime";
+import {
+  BedrockAgentRuntimeClient,
+  type BedrockAgentRuntimeClientConfig,
+  type RetrievalFilter,
+  RetrieveCommand,
+  type SearchType,
+} from "@aws-sdk/client-bedrock-agent-runtime";
+import type { QueryBundle } from "@llamaindex/core/query-engine";
+import { BaseRetriever } from "@llamaindex/core/retriever";
+import { Document, type NodeWithScore } from "@llamaindex/core/schema";
+import { extractText } from "@llamaindex/core/utils";
+
+/**
+ * Interface for the arguments required to initialize an
+ * AmazonKnowledgeBaseRetriever instance.
+ */
+export interface AmazonKnowledgeBaseRetrieverArgs {
+  knowledgeBaseId: string;
+  topK: number;
+  region: string;
+  clientOptions?: BedrockAgentRuntimeClientConfig;
+  filter?: RetrievalFilter;
+  overrideSearchType?: SearchType;
+}
+
+/**
+ * Class for interacting with Amazon Bedrock Knowledge Bases, a RAG workflow oriented service
+ * Extends the BaseRetriever class.
+ * @example
+ * ```typescript
+ * const retriever = new AmazonKnowledgeBaseRetriever({
+ *   topK: 10,
+ *   knowledgeBaseId: "YOUR_KNOWLEDGE_BASE_ID",
+ *   region: "us-east-2",
+ *   clientOptions: {
+ *     credentials: {
+ *       accessKeyId: "YOUR_ACCESS_KEY_ID",
+ *       secretAccessKey: "YOUR_SECRET_ACCESS_KEY",
+ *     },
+ *   },
+ * });
+ *
+ * const docs = await retriever.retrieve({query: "How are clouds formed?"});
+ * ```
+ */
+export class AmazonKnowledgeBaseRetriever extends BaseRetriever {
+  static lc_name() {
+    return "AmazonKnowledgeBaseRetriever";
+  }
+
+  lc_namespace = ["llamaindex", "retrievers", "amazon_bedrock_knowledge_base"];
+
+  knowledgeBaseId: string;
+
+  topK: number;
+
+  bedrockAgentRuntimeClient: BedrockAgentRuntimeClient;
+
+  filter: RetrievalFilter | undefined;
+
+  overrideSearchType: SearchType | undefined;
+
+  constructor({
+    knowledgeBaseId,
+    topK = 10,
+    clientOptions,
+    region,
+    filter,
+    overrideSearchType,
+  }: AmazonKnowledgeBaseRetrieverArgs) {
+    super();
+
+    this.topK = topK;
+    this.filter = filter;
+    this.overrideSearchType = overrideSearchType;
+    this.bedrockAgentRuntimeClient = new BedrockAgentRuntimeClient({
+      region,
+      ...clientOptions,
+    });
+    this.knowledgeBaseId = knowledgeBaseId;
+  }
+
+  /**
+   * Cleans the result text by replacing sequences of whitespace with a
+   * single space and removing ellipses.
+   * @param resText The result text to clean.
+   * @returns The cleaned result text.
+   */
+  cleanResult(resText: string) {
+    const res = resText.replace(/\s+/g, " ").replace(/\.\.\./g, "");
+    return res;
+  }
+
+  async queryKnowledgeBase(
+    query: QueryBundle,
+    topK: number,
+    filter?: RetrievalFilter,
+    overrideSearchType?: SearchType,
+  ): Promise<NodeWithScore[]> {
+    const retrieveCommand = new RetrieveCommand({
+      knowledgeBaseId: this.knowledgeBaseId,
+      retrievalQuery: {
+        text: extractText(query),
+      },
+      retrievalConfiguration: {
+        vectorSearchConfiguration: {
+          numberOfResults: topK,
+          overrideSearchType,
+          filter,
+        } as KnowledgeBaseVectorSearchConfiguration,
+      },
+    });
+
+    const retrieveResponse =
+      await this.bedrockAgentRuntimeClient.send(retrieveCommand);
+
+    return (
+      retrieveResponse.retrievalResults?.map((result) => {
+        let source;
+        switch (result.location?.type) {
+          case "CONFLUENCE":
+            source = result.location?.confluenceLocation?.url;
+            break;
+          case "S3":
+            source = result.location?.s3Location?.uri;
+            break;
+          case "SALESFORCE":
+            source = result.location?.salesforceLocation?.url;
+            break;
+          case "SHAREPOINT":
+            source = result.location?.sharePointLocation?.url;
+            break;
+          case "WEB":
+            source = result.location?.webLocation?.url;
+            break;
+          default:
+            source = result.location?.s3Location?.uri;
+            break;
+        }
+
+        return {
+          node: new Document({
+            text: this.cleanResult(result.content?.text || ""),
+            metadata: {
+              source,
+              score: result.score,
+              ...result.metadata,
+            },
+          }),
+          score: result.score ?? 1.0,
+        };
+      }) ?? []
+    );
+  }
+
+  async _retrieve(query: QueryBundle): Promise<NodeWithScore[]> {
+    return await this.queryKnowledgeBase(
+      query,
+      this.topK,
+      this.filter,
+      this.overrideSearchType,
+    );
+  }
+}

packages/core/CHANGELOG.md

@@ -1,5 +1,14 @@
 # @llamaindex/core
 
+## 0.2.6
+
+### Patch Changes
+
+- 8b7fdba: refactor: move chat engine & retriever into core.
+
+  - `chatHistory` in BaseChatEngine now returns `ChatMessage[] | Promise<ChatMessage[]>`, instead of `BaseMemory`
+  - update `retrieve-end` type
+
 ## 0.2.5
 
 ### Patch Changes

packages/core/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/core",
   "type": "module",
-  "version": "0.2.5",
+  "version": "0.2.6",
   "description": "LlamaIndex Core Module",
   "exports": {
     "./node-parser": {
@@ -199,6 +199,34 @@
         "types": "./dist/response-synthesizers/index.d.ts",
         "default": "./dist/response-synthesizers/index.js"
       }
+    },
+    "./chat-engine": {
+      "require": {
+        "types": "./dist/chat-engine/index.d.cts",
+        "default": "./dist/chat-engine/index.cjs"
+      },
+      "import": {
+        "types": "./dist/chat-engine/index.d.ts",
+        "default": "./dist/chat-engine/index.js"
+      },
+      "default": {
+        "types": "./dist/chat-engine/index.d.ts",
+        "default": "./dist/chat-engine/index.js"
+      }
+    },
+    "./retriever": {
+      "require": {
+        "types": "./dist/retriever/index.d.cts",
+        "default": "./dist/retriever/index.cjs"
+      },
+      "import": {
+        "types": "./dist/retriever/index.d.ts",
+        "default": "./dist/retriever/index.js"
+      },
+      "default": {
+        "types": "./dist/retriever/index.d.ts",
+        "default": "./dist/retriever/index.js"
+      }
     }
   },
   "files": [

packages/core/src/chat-engine/index.ts

@@ -0,0 +1,36 @@
+import type { ChatMessage, MessageContent } from "../llms";
+import type { BaseMemory } from "../memory";
+import { EngineResponse } from "../schema";
+
+export interface BaseChatEngineParams<
+  AdditionalMessageOptions extends object = object,
+> {
+  message: MessageContent;
+  /**
+   * Optional chat history if you want to customize the chat history.
+   */
+  chatHistory?:
+    | ChatMessage<AdditionalMessageOptions>[]
+    | BaseMemory<AdditionalMessageOptions>;
+}
+
+export interface StreamingChatEngineParams<
+  AdditionalMessageOptions extends object = object,
+> extends BaseChatEngineParams<AdditionalMessageOptions> {
+  stream: true;
+}
+
+export interface NonStreamingChatEngineParams<
+  AdditionalMessageOptions extends object = object,
+> extends BaseChatEngineParams<AdditionalMessageOptions> {
+  stream?: false;
+}
+
+export abstract class BaseChatEngine {
+  abstract chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
+  abstract chat(
+    params: StreamingChatEngineParams,
+  ): Promise<AsyncIterable<EngineResponse>>;
+
+  abstract chatHistory: ChatMessage[] | Promise<ChatMessage[]>;
+}

packages/core/src/global/settings/callback-manager.ts

@@ -11,6 +11,7 @@ import type {
   SynthesizeEndEvent,
   SynthesizeStartEvent,
 } from "../../response-synthesizers";
+import type { RetrieveEndEvent, RetrieveStartEvent } from "../../retriever";
 import { TextNode } from "../../schema";
 import { EventCaller, getEventCaller } from "../../utils";
 import type { UUID } from "../type";
@@ -69,6 +70,8 @@ export interface LlamaIndexEventMaps {
   "query-end": QueryEndEvent;
   "synthesize-start": SynthesizeStartEvent;
   "synthesize-end": SynthesizeEndEvent;
+  "retrieve-start": RetrieveStartEvent;
+  "retrieve-end": RetrieveEndEvent;
 }
 
 export class LlamaIndexCustomEvent<T = any> extends CustomEvent<T> {

packages/core/src/query-engine/base.ts

@@ -2,7 +2,7 @@ import { randomUUID } from "@llamaindex/env";
 import { Settings } from "../global";
 import type { MessageContent } from "../llms";
 import { PromptMixin } from "../prompts";
-import { EngineResponse } from "../schema";
+import { EngineResponse, type NodeWithScore } from "../schema";
 import { wrapEventCaller } from "../utils";
 
 /**
@@ -18,6 +18,18 @@ export type QueryBundle = {
 
 export type QueryType = string | QueryBundle;
 
+export type BaseQueryParams = {
+  query: QueryType;
+};
+
+export interface StreamingQueryParams extends BaseQueryParams {
+  stream: true;
+}
+
+export interface NonStreamingQueryParams extends BaseQueryParams {
+  stream?: false;
+}
+
 export type QueryFn = (
   strOrQueryBundle: QueryType,
   stream?: boolean,
@@ -28,23 +40,26 @@ export abstract class BaseQueryEngine extends PromptMixin {
     super();
   }
 
-  query(
-    strOrQueryBundle: QueryType,
-    stream: true,
-  ): Promise<AsyncIterable<EngineResponse>>;
-  query(strOrQueryBundle: QueryType, stream?: false): Promise<EngineResponse>;
+  async retrieve(params: QueryType): Promise<NodeWithScore[]> {
+    throw new Error(
+      "This query engine does not support retrieve, use query directly",
+    );
+  }
+
+  query(params: StreamingQueryParams): Promise<AsyncIterable<EngineResponse>>;
+  query(params: NonStreamingQueryParams): Promise<EngineResponse>;
   @wrapEventCaller
   async query(
-    strOrQueryBundle: QueryType,
-    stream = false,
+    params: StreamingQueryParams | NonStreamingQueryParams,
   ): Promise<EngineResponse | AsyncIterable<EngineResponse>> {
+    const { stream, query } = params;
     const id = randomUUID();
     const callbackManager = Settings.callbackManager;
     callbackManager.dispatchEvent("query-start", {
       id,
-      query: strOrQueryBundle,
+      query,
     });
-    const response = await this._query(strOrQueryBundle, stream);
+    const response = await this._query(query, stream);
     callbackManager.dispatchEvent("query-end", {
       id,
       response,

packages/core/src/retriever/index.ts

@@ -0,0 +1,112 @@
+import { randomUUID } from "@llamaindex/env";
+import { Settings } from "../global";
+import type { MessageContent } from "../llms";
+import { PromptMixin } from "../prompts";
+import type { QueryBundle, QueryType } from "../query-engine";
+import { BaseNode, IndexNode, type NodeWithScore, ObjectType } from "../schema";
+
+export type RetrieveParams = {
+  query: MessageContent;
+  preFilters?: unknown;
+};
+
+export type RetrieveStartEvent = {
+  id: string;
+  query: QueryBundle;
+};
+
+export type RetrieveEndEvent = {
+  id: string;
+  query: QueryBundle;
+  nodes: NodeWithScore[];
+};
+
+export abstract class BaseRetriever extends PromptMixin {
+  objectMap: Map<string, unknown> = new Map();
+
+  protected _updatePrompts() {}
+  protected _getPrompts() {
+    return {};
+  }
+
+  protected _getPromptModules() {
+    return {};
+  }
+
+  protected constructor() {
+    super();
+  }
+
+  public async retrieve(params: QueryType): Promise<NodeWithScore[]> {
+    const cb = Settings.callbackManager;
+    const queryBundle = typeof params === "string" ? { query: params } : params;
+    const id = randomUUID();
+    cb.dispatchEvent("retrieve-start", { id, query: queryBundle });
+    let response = await this._retrieve(queryBundle);
+    response = await this._handleRecursiveRetrieval(queryBundle, response);
+    cb.dispatchEvent("retrieve-end", {
+      id,
+      query: queryBundle,
+      nodes: response,
+    });
+    return response;
+  }
+
+  abstract _retrieve(params: QueryBundle): Promise<NodeWithScore[]>;
+
+  async _handleRecursiveRetrieval(
+    params: QueryBundle,
+    nodes: NodeWithScore[],
+  ): Promise<NodeWithScore[]> {
+    const retrievedNodes = [];
+    for (const { node, score = 1.0 } of nodes) {
+      if (node.type === ObjectType.INDEX) {
+        const indexNode = node as IndexNode;
+        const object = this.objectMap.get(indexNode.indexId);
+        if (object !== undefined) {
+          retrievedNodes.push(
+            ...this._retrieveFromObject(object, params, score),
+          );
+        } else {
+          retrievedNodes.push({ node, score });
+        }
+      } else {
+        retrievedNodes.push({ node, score });
+      }
+    }
+    return nodes;
+  }
+
+  _retrieveFromObject(
+    object: unknown,
+    queryBundle: QueryBundle,
+    score: number,
+  ): NodeWithScore[] {
+    if (object == null) {
+      throw new TypeError("Object is not retrievable");
+    }
+    if (typeof object !== "object") {
+      throw new TypeError("Object is not retrievable");
+    }
+    if ("node" in object && object.node instanceof BaseNode) {
+      return [
+        {
+          node: object.node,
+          score:
+            "score" in object && typeof object.score === "number"
+              ? object.score
+              : score,
+        },
+      ];
+    }
+    if (object instanceof BaseNode) {
+      return [{ node: object, score }];
+    } else {
+      // todo: support other types
+      // BaseQueryEngine
+      // BaseRetriever
+      // QueryComponent
+      throw new TypeError("Object is not retrievable");
+    }
+  }
+}

packages/experimental/CHANGELOG.md

@@ -1,5 +1,25 @@
 # @llamaindex/experimental
 
+## 0.0.87
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.0.86
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.0.85
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.0.84
 
 ### Patch Changes

packages/experimental/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/experimental",
   "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.84",
+  "version": "0.0.87",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/llamaindex/CHANGELOG.md

@@ -1,5 +1,34 @@
 # llamaindex
 
+## 0.6.9
+
+### Patch Changes
+
+- Updated dependencies [ac41ed3]
+  - @llamaindex/cloud@0.2.8
+
+## 0.6.8
+
+### Patch Changes
+
+- 8b7fdba: refactor: move chat engine & retriever into core.
+
+  - `chatHistory` in BaseChatEngine now returns `ChatMessage[] | Promise<ChatMessage[]>`, instead of `BaseMemory`
+  - update `retrieve-end` type
+
+- Updated dependencies [8b7fdba]
+  - @llamaindex/core@0.2.6
+  - @llamaindex/openai@0.1.8
+  - @llamaindex/groq@0.0.7
+
+## 0.6.7
+
+### Patch Changes
+
+- 23bcc37: fix: add `serializer` in doc store
+
+  `PostgresDocumentStore` now will not use JSON.stringify for better performance
+
 ## 0.6.6
 
 ### Patch Changes

packages/llamaindex/e2e/examples/cloudflare-worker-agent/CHANGELOG.md

@@ -1,5 +1,25 @@
 # @llamaindex/cloudflare-worker-agent-test
 
+## 0.0.71
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.0.70
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.0.69
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.0.68
 
 ### Patch Changes

packages/llamaindex/e2e/examples/cloudflare-worker-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.68",
+  "version": "0.0.71",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/e2e/examples/llama-parse-browser/CHANGELOG.md

@@ -1,5 +1,12 @@
 # @llamaindex/llama-parse-browser-test
 
+## 0.0.4
+
+### Patch Changes
+
+- Updated dependencies [ac41ed3]
+  - @llamaindex/cloud@0.2.8
+
 ## 0.0.3
 
 ### Patch Changes

packages/llamaindex/e2e/examples/llama-parse-browser/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/llama-parse-browser-test",
   "private": true,
-  "version": "0.0.3",
+  "version": "0.0.4",
   "type": "module",
   "scripts": {
     "dev": "vite",

packages/llamaindex/e2e/examples/nextjs-agent/CHANGELOG.md

@@ -1,5 +1,25 @@
 # @llamaindex/next-agent-test
 
+## 0.1.71
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.1.70
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.1.69
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.1.68
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-agent-test",
-  "version": "0.1.68",
+  "version": "0.1.71",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-edge-runtime/CHANGELOG.md

@@ -1,5 +1,25 @@
 # test-edge-runtime
 
+## 0.1.70
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.1.69
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.1.68
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.1.67
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-edge-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.67",
+  "version": "0.1.70",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-node-runtime/CHANGELOG.md

@@ -1,5 +1,25 @@
 # @llamaindex/next-node-runtime
 
+## 0.0.52
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.0.51
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.0.50
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.0.49
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-node-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.0.49",
+  "version": "0.0.52",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/waku-query-engine/CHANGELOG.md

@@ -1,5 +1,25 @@
 # @llamaindex/waku-query-engine-test
 
+## 0.0.71
+
+### Patch Changes
+
+- llamaindex@0.6.9
+
+## 0.0.70
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - llamaindex@0.6.8
+
+## 0.0.69
+
+### Patch Changes
+
+- Updated dependencies [23bcc37]
+  - llamaindex@0.6.7
+
 ## 0.0.68
 
 ### Patch Changes

packages/llamaindex/e2e/examples/waku-query-engine/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.68",
+  "version": "0.0.71",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llamaindex",
-  "version": "0.6.6",
+  "version": "0.6.9",
   "license": "MIT",
   "type": "module",
   "keywords": [

packages/llamaindex/src/Retriever.ts

@@ -1,20 +0,0 @@
-import type { NodeWithScore } from "@llamaindex/core/schema";
-import type { ServiceContext } from "./ServiceContext.js";
-import type { MessageContent } from "./index.edge.js";
-
-export type RetrieveParams = {
-  query: MessageContent;
-  preFilters?: unknown;
-};
-
-/**
- * Retrievers retrieve the nodes that most closely match our query in similarity.
- */
-export interface BaseRetriever {
-  retrieve(params: RetrieveParams): Promise<NodeWithScore[]>;
-
-  /**
-   * @deprecated to be deprecated soon
-   */
-  serviceContext?: ServiceContext | undefined;
-}

packages/llamaindex/src/agent/anthropic.ts

@@ -1,9 +1,9 @@
-import { Settings } from "../Settings.js";
 import type {
-  ChatEngineParamsNonStreaming,
-  ChatEngineParamsStreaming,
-  EngineResponse,
-} from "../index.edge.js";
+  NonStreamingChatEngineParams,
+  StreamingChatEngineParams,
+} from "@llamaindex/core/chat-engine";
+import type { EngineResponse } from "@llamaindex/core/schema";
+import { Settings } from "../Settings.js";
 import { Anthropic } from "../llm/anthropic.js";
 import { LLMAgent, LLMAgentWorker, type LLMAgentParams } from "./llm.js";
 
@@ -24,12 +24,13 @@ export class AnthropicAgent extends LLMAgent {
     });
   }
 
-  async chat(params: ChatEngineParamsNonStreaming): Promise<EngineResponse>;
-  async chat(params: ChatEngineParamsStreaming): Promise<never>;
+  async chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
+  async chat(params: StreamingChatEngineParams): Promise<never>;
   override async chat(
-    params: ChatEngineParamsNonStreaming | ChatEngineParamsStreaming,
+    params: NonStreamingChatEngineParams | StreamingChatEngineParams,
   ) {
-    if (params.stream) {
+    const { stream } = params;
+    if (stream) {
       // Anthropic does support this, but looks like it's not supported in the LITS LLM
       throw new Error("Anthropic does not support streaming");
     }

packages/llamaindex/src/agent/base.ts

@@ -1,3 +1,8 @@
+import {
+  BaseChatEngine,
+  type NonStreamingChatEngineParams,
+  type StreamingChatEngineParams,
+} from "@llamaindex/core/chat-engine";
 import type {
   BaseToolWithCall,
   ChatMessage,
@@ -10,11 +15,6 @@ import { EngineResponse } from "@llamaindex/core/schema";
 import { wrapEventCaller } from "@llamaindex/core/utils";
 import { randomUUID } from "@llamaindex/env";
 import { Settings } from "../Settings.js";
-import {
-  type ChatEngine,
-  type ChatEngineParamsNonStreaming,
-  type ChatEngineParamsStreaming,
-} from "../engines/chat/index.js";
 import { consoleLogger, emptyLogger } from "../internal/logger.js";
 import { isReadableStream } from "../internal/utils.js";
 import { ObjectRetriever } from "../objects/index.js";
@@ -207,8 +207,7 @@ export abstract class AgentRunner<
   >
     ? AdditionalMessageOptions
     : never,
-> implements ChatEngine
-{
+> extends BaseChatEngine {
   readonly #llm: AI;
   readonly #tools:
     | BaseToolWithCall[]
@@ -259,6 +258,7 @@ export abstract class AgentRunner<
   protected constructor(
     params: AgentRunnerParams<AI, Store, AdditionalMessageOptions>,
   ) {
+    super();
     const { llm, chatHistory, systemPrompt, runner, tools, verbose } = params;
     this.#llm = llm;
     this.#chatHistory = chatHistory;
@@ -345,13 +345,13 @@ export abstract class AgentRunner<
     });
   }
 
-  async chat(params: ChatEngineParamsNonStreaming): Promise<EngineResponse>;
+  async chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
   async chat(
-    params: ChatEngineParamsStreaming,
+    params: StreamingChatEngineParams,
   ): Promise<ReadableStream<EngineResponse>>;
   @wrapEventCaller
   async chat(
-    params: ChatEngineParamsNonStreaming | ChatEngineParamsStreaming,
+    params: NonStreamingChatEngineParams | StreamingChatEngineParams,
   ): Promise<EngineResponse | ReadableStream<EngineResponse>> {
     let chatHistory: ChatMessage<AdditionalMessageOptions>[] = [];
 

packages/llamaindex/src/cloud/LlamaCloudIndex.ts

@@ -1,7 +1,6 @@
 import type { BaseQueryEngine } from "@llamaindex/core/query-engine";
 import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
 import type { Document, TransformComponent } from "@llamaindex/core/schema";
-import type { BaseRetriever } from "../Retriever.js";
 import { RetrieverQueryEngine } from "../engines/query/RetrieverQueryEngine.js";
 import type { BaseNodePostprocessor } from "../postprocessors/types.js";
 import type { CloudRetrieveParams } from "./LlamaCloudRetriever.js";
@@ -12,6 +11,7 @@ import { getAppBaseUrl, getProjectId, initService } from "./utils.js";
 
 import { PipelinesService, ProjectsService } from "@llamaindex/cloud/api";
 import { SentenceSplitter } from "@llamaindex/core/node-parser";
+import type { BaseRetriever } from "@llamaindex/core/retriever";
 import { getEnv } from "@llamaindex/env";
 import { OpenAIEmbedding } from "@llamaindex/openai";
 import { Settings } from "../Settings.js";

packages/llamaindex/src/cloud/LlamaCloudRetriever.ts

@@ -4,11 +4,12 @@ import {
   type RetrievalParams,
   type TextNodeWithScore,
 } from "@llamaindex/cloud/api";
-import { DEFAULT_PROJECT_NAME, Settings } from "@llamaindex/core/global";
+import { DEFAULT_PROJECT_NAME } from "@llamaindex/core/global";
+import type { QueryBundle } from "@llamaindex/core/query-engine";
+import { BaseRetriever } from "@llamaindex/core/retriever";
 import type { NodeWithScore } from "@llamaindex/core/schema";
 import { jsonToNode, ObjectType } from "@llamaindex/core/schema";
-import { extractText, wrapEventCaller } from "@llamaindex/core/utils";
-import type { BaseRetriever, RetrieveParams } from "../Retriever.js";
+import { extractText } from "@llamaindex/core/utils";
 import type { ClientParams, CloudConstructorParams } from "./type.js";
 import { getProjectId, initService } from "./utils.js";
 
@@ -17,7 +18,7 @@ export type CloudRetrieveParams = Omit<
   "query" | "search_filters" | "dense_similarity_top_k"
 > & { similarityTopK?: number; filters?: MetadataFilters };
 
-export class LlamaCloudRetriever implements BaseRetriever {
+export class LlamaCloudRetriever extends BaseRetriever {
   clientParams: ClientParams;
   retrieveParams: CloudRetrieveParams;
   organizationId?: string;
@@ -36,12 +37,13 @@ export class LlamaCloudRetriever implements BaseRetriever {
       return {
         // Currently LlamaCloud only supports text nodes
         node: textNode,
-        score: node.score,
+        score: node.score ?? undefined,
       };
     });
   }
 
   constructor(params: CloudConstructorParams & CloudRetrieveParams) {
+    super();
     this.clientParams = { apiKey: params.apiKey, baseUrl: params.baseUrl };
     initService(this.clientParams);
     this.retrieveParams = params;
@@ -54,11 +56,7 @@ export class LlamaCloudRetriever implements BaseRetriever {
     }
   }
 
-  @wrapEventCaller
-  async retrieve({
-    query,
-    preFilters,
-  }: RetrieveParams): Promise<NodeWithScore[]> {
+  async _retrieve(query: QueryBundle): Promise<NodeWithScore[]> {
     const { data: pipelines } =
       await PipelinesService.searchPipelinesApiV1PipelinesGet({
         query: {
@@ -97,19 +95,11 @@ export class LlamaCloudRetriever implements BaseRetriever {
         body: {
           ...this.retrieveParams,
           query: extractText(query),
-          search_filters:
-            this.retrieveParams.filters ?? (preFilters as MetadataFilters),
+          search_filters: this.retrieveParams.filters as MetadataFilters,
           dense_similarity_top_k: this.retrieveParams.similarityTopK!,
         },
       });
 
-    const nodesWithScores = this.resultNodesToNodeWithScore(
-      results.retrieval_nodes,
-    );
-    Settings.callbackManager.dispatchEvent("retrieve-end", {
-      query,
-      nodes: nodesWithScores,
-    });
-    return nodesWithScores;
+    return this.resultNodesToNodeWithScore(results.retrieval_nodes);
   }
 }

packages/llamaindex/src/engines/chat/CondenseQuestionChatEngine.ts

@@ -1,10 +1,14 @@
+import {
+  BaseChatEngine,
+  type NonStreamingChatEngineParams,
+  type StreamingChatEngineParams,
+} from "@llamaindex/core/chat-engine";
 import type { ChatMessage, LLM } from "@llamaindex/core/llms";
 import { BaseMemory, ChatMemoryBuffer } from "@llamaindex/core/memory";
 import {
   type CondenseQuestionPrompt,
   defaultCondenseQuestionPrompt,
   type ModuleRecord,
-  PromptMixin,
 } from "@llamaindex/core/prompts";
 import type { BaseQueryEngine } from "@llamaindex/core/query-engine";
 import type { EngineResponse } from "@llamaindex/core/schema";
@@ -16,11 +20,6 @@ import {
 } from "@llamaindex/core/utils";
 import type { ServiceContext } from "../../ServiceContext.js";
 import { llmFromSettingsOrContext } from "../../Settings.js";
-import type {
-  ChatEngine,
-  ChatEngineParamsNonStreaming,
-  ChatEngineParamsStreaming,
-} from "./types.js";
 
 /**
  * CondenseQuestionChatEngine is used in conjunction with a Index (for example VectorStoreIndex).
@@ -32,16 +31,16 @@ import type {
  * underlying data. It performs less well when the chat messages are not questions about the
  * data, or are very referential to previous context.
  */
-
-export class CondenseQuestionChatEngine
-  extends PromptMixin
-  implements ChatEngine
-{
+export class CondenseQuestionChatEngine extends BaseChatEngine {
   queryEngine: BaseQueryEngine;
-  chatHistory: BaseMemory;
+  memory: BaseMemory;
   llm: LLM;
   condenseMessagePrompt: CondenseQuestionPrompt;
 
+  get chatHistory() {
+    return this.memory.getMessages();
+  }
+
   constructor(init: {
     queryEngine: BaseQueryEngine;
     chatHistory: ChatMessage[];
@@ -51,7 +50,7 @@ export class CondenseQuestionChatEngine
     super();
 
     this.queryEngine = init.queryEngine;
-    this.chatHistory = new ChatMemoryBuffer({
+    this.memory = new ChatMemoryBuffer({
       chatHistory: init?.chatHistory,
     });
     this.llm = llmFromSettingsOrContext(init?.serviceContext);
@@ -88,13 +87,13 @@ export class CondenseQuestionChatEngine
     });
   }
 
+  chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
   chat(
-    params: ChatEngineParamsStreaming,
+    params: StreamingChatEngineParams,
   ): Promise<AsyncIterable<EngineResponse>>;
-  chat(params: ChatEngineParamsNonStreaming): Promise<EngineResponse>;
   @wrapEventCaller
   async chat(
-    params: ChatEngineParamsStreaming | ChatEngineParamsNonStreaming,
+    params: NonStreamingChatEngineParams | StreamingChatEngineParams,
   ): Promise<EngineResponse | AsyncIterable<EngineResponse>> {
     const { message, stream } = params;
     const chatHistory = params.chatHistory
@@ -104,7 +103,7 @@ export class CondenseQuestionChatEngine
               ? await params.chatHistory.getMessages()
               : params.chatHistory,
         })
-      : this.chatHistory;
+      : this.memory;
 
     const condensedQuestion = (
       await this.condenseQuestion(chatHistory, extractText(message))
@@ -112,12 +111,10 @@ export class CondenseQuestionChatEngine
     chatHistory.put({ content: message, role: "user" });
 
     if (stream) {
-      const stream = await this.queryEngine.query(
-        {
-          query: condensedQuestion,
-        },
-        true,
-      );
+      const stream = await this.queryEngine.query({
+        query: condensedQuestion,
+        stream: true,
+      });
       return streamReducer({
         stream,
         initialValue: "",
@@ -140,6 +137,6 @@ export class CondenseQuestionChatEngine
   }
 
   reset() {
-    this.chatHistory.reset();
+    this.memory.reset();
   }
 }

packages/llamaindex/src/engines/chat/ContextChatEngine.ts

@@ -1,3 +1,8 @@
+import type {
+  BaseChatEngine,
+  NonStreamingChatEngineParams,
+  StreamingChatEngineParams,
+} from "@llamaindex/core/chat-engine";
 import type {
   ChatMessage,
   LLM,
@@ -11,6 +16,7 @@ import {
   PromptMixin,
   type PromptsRecord,
 } from "@llamaindex/core/prompts";
+import type { BaseRetriever } from "@llamaindex/core/retriever";
 import { EngineResponse, MetadataMode } from "@llamaindex/core/schema";
 import {
   extractText,
@@ -18,27 +24,25 @@ import {
   streamReducer,
   wrapEventCaller,
 } from "@llamaindex/core/utils";
-import type { BaseRetriever } from "../../Retriever.js";
 import { Settings } from "../../Settings.js";
 import type { BaseNodePostprocessor } from "../../postprocessors/index.js";
 import { DefaultContextGenerator } from "./DefaultContextGenerator.js";
-import type {
-  ChatEngine,
-  ChatEngineParamsNonStreaming,
-  ChatEngineParamsStreaming,
-  ContextGenerator,
-} from "./types.js";
+import type { ContextGenerator } from "./types.js";
 
 /**
  * ContextChatEngine uses the Index to get the appropriate context for each query.
  * The context is stored in the system prompt, and the chat history is chunk: ChatResponseChunk, nodes?: NodeWithScore<import("/Users/marcus/code/llamaindex/LlamaIndexTS/packages/core/src/Node").Metadata>[], nodes?: NodeWithScore<import("/Users/marcus/code/llamaindex/LlamaIndexTS/packages/core/src/Node").Metadata>[]lowing the appropriate context to be surfaced for each query.
  */
-export class ContextChatEngine extends PromptMixin implements ChatEngine {
+export class ContextChatEngine extends PromptMixin implements BaseChatEngine {
   chatModel: LLM;
-  chatHistory: BaseMemory;
+  memory: BaseMemory;
   contextGenerator: ContextGenerator & PromptMixin;
   systemPrompt?: string | undefined;
 
+  get chatHistory() {
+    return this.memory.getMessages();
+  }
+
   constructor(init: {
     retriever: BaseRetriever;
     chatModel?: LLM | undefined;
@@ -50,7 +54,7 @@ export class ContextChatEngine extends PromptMixin implements ChatEngine {
   }) {
     super();
     this.chatModel = init.chatModel ?? Settings.llm;
-    this.chatHistory = new ChatMemoryBuffer({ chatHistory: init?.chatHistory });
+    this.memory = new ChatMemoryBuffer({ chatHistory: init?.chatHistory });
     this.contextGenerator = new DefaultContextGenerator({
       retriever: init.retriever,
       contextSystemPrompt: init?.contextSystemPrompt,
@@ -79,13 +83,13 @@ export class ContextChatEngine extends PromptMixin implements ChatEngine {
     };
   }
 
+  chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
   chat(
-    params: ChatEngineParamsStreaming,
+    params: StreamingChatEngineParams,
   ): Promise<AsyncIterable<EngineResponse>>;
-  chat(params: ChatEngineParamsNonStreaming): Promise<EngineResponse>;
   @wrapEventCaller
   async chat(
-    params: ChatEngineParamsStreaming | ChatEngineParamsNonStreaming,
+    params: StreamingChatEngineParams | NonStreamingChatEngineParams,
   ): Promise<EngineResponse | AsyncIterable<EngineResponse>> {
     const { message, stream } = params;
     const chatHistory = params.chatHistory
@@ -95,7 +99,7 @@ export class ContextChatEngine extends PromptMixin implements ChatEngine {
               ? await params.chatHistory.getMessages()
               : params.chatHistory,
         })
-      : this.chatHistory;
+      : this.memory;
     const requestMessages = await this.prepareRequestMessages(
       message,
       chatHistory,
@@ -125,7 +129,7 @@ export class ContextChatEngine extends PromptMixin implements ChatEngine {
   }
 
   reset() {
-    this.chatHistory.reset();
+    this.memory.reset();
   }
 
   private async prepareRequestMessages(

packages/llamaindex/src/engines/chat/DefaultContextGenerator.ts

@@ -5,10 +5,10 @@ import {
   type ModuleRecord,
   PromptMixin,
 } from "@llamaindex/core/prompts";
+import type { BaseRetriever } from "@llamaindex/core/retriever";
 import { MetadataMode, type NodeWithScore } from "@llamaindex/core/schema";
 import { createMessageContent } from "@llamaindex/core/utils";
 import type { BaseNodePostprocessor } from "../../postprocessors/index.js";
-import type { BaseRetriever } from "../../Retriever.js";
 import type { Context, ContextGenerator } from "./types.js";
 
 export class DefaultContextGenerator

packages/llamaindex/src/engines/chat/SimpleChatEngine.ts

@@ -1,3 +1,8 @@
+import type {
+  BaseChatEngine,
+  NonStreamingChatEngineParams,
+  StreamingChatEngineParams,
+} from "@llamaindex/core/chat-engine";
 import type { LLM } from "@llamaindex/core/llms";
 import { BaseMemory, ChatMemoryBuffer } from "@llamaindex/core/memory";
 import { EngineResponse } from "@llamaindex/core/schema";
@@ -7,32 +12,31 @@ import {
   wrapEventCaller,
 } from "@llamaindex/core/utils";
 import { Settings } from "../../Settings.js";
-import type {
-  ChatEngine,
-  ChatEngineParamsNonStreaming,
-  ChatEngineParamsStreaming,
-} from "./types.js";
 
 /**
  * SimpleChatEngine is the simplest possible chat engine. Useful for using your own custom prompts.
  */
 
-export class SimpleChatEngine implements ChatEngine {
-  chatHistory: BaseMemory;
+export class SimpleChatEngine implements BaseChatEngine {
+  memory: BaseMemory;
   llm: LLM;
 
+  get chatHistory() {
+    return this.memory.getMessages();
+  }
+
   constructor(init?: Partial<SimpleChatEngine>) {
-    this.chatHistory = init?.chatHistory ?? new ChatMemoryBuffer();
+    this.memory = init?.memory ?? new ChatMemoryBuffer();
     this.llm = init?.llm ?? Settings.llm;
   }
 
+  chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
   chat(
-    params: ChatEngineParamsStreaming,
+    params: StreamingChatEngineParams,
   ): Promise<AsyncIterable<EngineResponse>>;
-  chat(params: ChatEngineParamsNonStreaming): Promise<EngineResponse>;
   @wrapEventCaller
   async chat(
-    params: ChatEngineParamsStreaming | ChatEngineParamsNonStreaming,
+    params: NonStreamingChatEngineParams | StreamingChatEngineParams,
   ): Promise<EngineResponse | AsyncIterable<EngineResponse>> {
     const { message, stream } = params;
 
@@ -43,7 +47,7 @@ export class SimpleChatEngine implements ChatEngine {
               ? await params.chatHistory.getMessages()
               : params.chatHistory,
         })
-      : this.chatHistory;
+      : this.memory;
     chatHistory.put({ content: message, role: "user" });
 
     if (stream) {
@@ -73,6 +77,6 @@ export class SimpleChatEngine implements ChatEngine {
   }
 
   reset() {
-    this.chatHistory.reset();
+    this.memory.reset();
   }
 }

packages/llamaindex/src/engines/chat/types.ts

@@ -1,58 +1,10 @@
-import type { ChatMessage, MessageContent } from "@llamaindex/core/llms";
-import type { BaseMemory } from "@llamaindex/core/memory";
-import { EngineResponse, type NodeWithScore } from "@llamaindex/core/schema";
-
-/**
- * Represents the base parameters for ChatEngine.
- */
-export interface ChatEngineParamsBase {
-  message: MessageContent;
-  /**
-   * Optional chat history if you want to customize the chat history.
-   */
-  chatHistory?: ChatMessage[] | BaseMemory;
-  /**
-   * Optional flag to enable verbose mode.
-   * @default false
-   */
-  verbose?: boolean;
-}
-
-export interface ChatEngineParamsStreaming extends ChatEngineParamsBase {
-  stream: true;
-}
-
-export interface ChatEngineParamsNonStreaming extends ChatEngineParamsBase {
-  stream?: false | null;
-}
-
-/**
- * A ChatEngine is used to handle back and forth chats between the application and the LLM.
- */
-export interface ChatEngine<
-  // synchronous response
-  R = EngineResponse,
-  // asynchronous response
-  AR extends AsyncIterable<unknown> = AsyncIterable<R>,
-> {
-  /**
-   * Send message along with the class's current chat history to the LLM.
-   * @param params
-   */
-  chat(params: ChatEngineParamsStreaming): Promise<AR>;
-  chat(params: ChatEngineParamsNonStreaming): Promise<R>;
-
-  /**
-   * Resets the chat history so that it's empty.
-   */
-  reset(): void;
-}
+import type { ChatMessage } from "@llamaindex/core/llms";
+import type { NodeWithScore } from "@llamaindex/core/schema";
 
 export interface Context {
   message: ChatMessage;
   nodes: NodeWithScore[];
 }
-
 /**
  * A ContextGenerator is used to generate a context based on a message's text content
  */

packages/llamaindex/src/engines/query/RetrieverQueryEngine.ts

@@ -1,10 +1,11 @@
-import { BaseQueryEngine } from "@llamaindex/core/query-engine";
+import type { MessageContent } from "@llamaindex/core/llms";
+import { BaseQueryEngine, type QueryType } from "@llamaindex/core/query-engine";
 import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
 import { getResponseSynthesizer } from "@llamaindex/core/response-synthesizers";
+import { BaseRetriever } from "@llamaindex/core/retriever";
 import { type NodeWithScore } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
 import type { BaseNodePostprocessor } from "../../postprocessors/index.js";
-import type { BaseRetriever } from "../../Retriever.js";
 
 /**
  * A query engine that uses a retriever to query an index and then synthesizes the response.
@@ -67,7 +68,10 @@ export class RetrieverQueryEngine extends BaseQueryEngine {
     };
   }
 
-  private async applyNodePostprocessors(nodes: NodeWithScore[], query: string) {
+  private async applyNodePostprocessors(
+    nodes: NodeWithScore[],
+    query: MessageContent,
+  ) {
     let nodesWithScore = nodes;
 
     for (const postprocessor of this.nodePostprocessors) {
@@ -80,12 +84,10 @@ export class RetrieverQueryEngine extends BaseQueryEngine {
     return nodesWithScore;
   }
 
-  private async retrieve(query: string) {
-    const nodes = await this.retriever.retrieve({
-      query,
-      preFilters: this.preFilters,
-    });
+  override async retrieve(query: QueryType) {
+    const nodes = await this.retriever.retrieve(query);
 
-    return await this.applyNodePostprocessors(nodes, query);
+    const messageContent = typeof query === "string" ? query : query.query;
+    return await this.applyNodePostprocessors(nodes, messageContent);
   }
 }

packages/llamaindex/src/engines/query/RouterQueryEngine.ts

@@ -136,7 +136,9 @@ export class RouterQueryEngine extends BaseQueryEngine {
         }
 
         const selectedQueryEngine = this.queryEngines[engineInd.index]!;
-        responses.push(await selectedQueryEngine.query(query));
+        responses.push(
+          await selectedQueryEngine.query({ query, stream: false }),
+        );
       }
 
       if (responses.length > 1) {

packages/llamaindex/src/evaluation/Faithfulness.ts

@@ -103,7 +103,8 @@ export class FaithfulnessEvaluator
     });
 
     const responseObj = await queryEngine.query({
-      query: response,
+      query: { query: response },
+      stream: false,
     });
 
     const rawResponseTxt = responseObj.toString();

packages/llamaindex/src/index.edge.ts

@@ -1,6 +1,6 @@
 import type { AgentEndEvent, AgentStartEvent } from "./agent/types.js";
-import type { RetrievalEndEvent, RetrievalStartEvent } from "./llm/types.js";
 
+export * from "@llamaindex/core/chat-engine";
 export {
   CallbackManager,
   DEFAULT_BASE_URL,
@@ -35,12 +35,11 @@ export * from "@llamaindex/core/llms";
 export * from "@llamaindex/core/prompts";
 export * from "@llamaindex/core/query-engine";
 export * from "@llamaindex/core/response-synthesizers";
+export * from "@llamaindex/core/retriever";
 export * from "@llamaindex/core/schema";
 
 declare module "@llamaindex/core/global" {
   export interface LlamaIndexEventMaps {
-    "retrieve-start": RetrievalStartEvent;
-    "retrieve-end": RetrievalEndEvent;
     // agent events
     "agent-start": AgentStartEvent;
     "agent-end": AgentEndEvent;
@@ -66,7 +65,6 @@ export * from "./objects/index.js";
 export * from "./OutputParser.js";
 export * from "./postprocessors/index.js";
 export * from "./QuestionGenerator.js";
-export * from "./Retriever.js";
 export * from "./selectors/index.js";
 export * from "./ServiceContext.js";
 export { Settings } from "./Settings.js";

packages/llamaindex/src/indices/BaseIndex.ts

@@ -1,7 +1,7 @@
 import type { BaseQueryEngine } from "@llamaindex/core/query-engine";
 import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
+import type { BaseRetriever } from "@llamaindex/core/retriever";
 import type { BaseNode, Document } from "@llamaindex/core/schema";
-import type { BaseRetriever } from "../Retriever.js";
 import type { ServiceContext } from "../ServiceContext.js";
 import { nodeParserFromSettingsOrContext } from "../Settings.js";
 import { runTransformations } from "../ingestion/IngestionPipeline.js";

packages/llamaindex/src/indices/keyword/index.ts

@@ -5,7 +5,6 @@ import type {
   NodeWithScore,
 } from "@llamaindex/core/schema";
 import { MetadataMode } from "@llamaindex/core/schema";
-import type { BaseRetriever, RetrieveParams } from "../../Retriever.js";
 import type { ServiceContext } from "../../ServiceContext.js";
 import { serviceContextFromDefaults } from "../../ServiceContext.js";
 import { RetrieverQueryEngine } from "../../engines/query/index.js";
@@ -29,7 +28,11 @@ import {
   type KeywordExtractPrompt,
   type QueryKeywordExtractPrompt,
 } from "@llamaindex/core/prompts";
-import type { BaseQueryEngine } from "@llamaindex/core/query-engine";
+import type {
+  BaseQueryEngine,
+  QueryBundle,
+} from "@llamaindex/core/query-engine";
+import { BaseRetriever } from "@llamaindex/core/retriever";
 import { extractText } from "@llamaindex/core/utils";
 import { llmFromSettingsOrContext } from "../../Settings.js";
 
@@ -48,7 +51,7 @@ export enum KeywordTableRetrieverMode {
 }
 
 // Base Keyword Table Retriever
-abstract class BaseKeywordTableRetriever implements BaseRetriever {
+abstract class BaseKeywordTableRetriever extends BaseRetriever {
   protected index: KeywordTableIndex;
   protected indexStruct: KeywordTable;
   protected docstore: BaseDocumentStore;
@@ -72,6 +75,7 @@ abstract class BaseKeywordTableRetriever implements BaseRetriever {
     maxKeywordsPerQuery: number;
     numChunksPerQuery: number;
   }) {
+    super();
     this.index = index;
     this.indexStruct = index.indexStruct;
     this.docstore = index.docStore;
@@ -87,7 +91,7 @@ abstract class BaseKeywordTableRetriever implements BaseRetriever {
 
   abstract getKeywords(query: string): Promise<string[]>;
 
-  async retrieve({ query }: RetrieveParams): Promise<NodeWithScore[]> {
+  async _retrieve(query: QueryBundle): Promise<NodeWithScore[]> {
     const keywords = await this.getKeywords(extractText(query));
     const chunkIndicesCount: { [key: string]: number } = {};
     const filteredKeywords = keywords.filter((keyword) =>

packages/llamaindex/src/indices/summary/index.ts

@@ -2,16 +2,17 @@ import {
   type ChoiceSelectPrompt,
   defaultChoiceSelectPrompt,
 } from "@llamaindex/core/prompts";
+import type { QueryBundle } from "@llamaindex/core/query-engine";
 import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
 import { getResponseSynthesizer } from "@llamaindex/core/response-synthesizers";
+import { BaseRetriever } from "@llamaindex/core/retriever";
 import type {
   BaseNode,
   Document,
   NodeWithScore,
 } from "@llamaindex/core/schema";
-import { extractText, wrapEventCaller } from "@llamaindex/core/utils";
+import { extractText } from "@llamaindex/core/utils";
 import _ from "lodash";
-import type { BaseRetriever, RetrieveParams } from "../../Retriever.js";
 import type { ServiceContext } from "../../ServiceContext.js";
 import {
   llmFromSettingsOrContext,
@@ -279,15 +280,15 @@ export type ListRetrieverMode = SummaryRetrieverMode;
 /**
  * Simple retriever for SummaryIndex that returns all nodes
  */
-export class SummaryIndexRetriever implements BaseRetriever {
+export class SummaryIndexRetriever extends BaseRetriever {
   index: SummaryIndex;
 
   constructor(index: SummaryIndex) {
+    super();
     this.index = index;
   }
 
-  @wrapEventCaller
-  async retrieve({ query }: RetrieveParams): Promise<NodeWithScore[]> {
+  async _retrieve(queryBundle: QueryBundle): Promise<NodeWithScore[]> {
     const nodeIds = this.index.indexStruct.nodes;
     const nodes = await this.index.docStore.getNodes(nodeIds);
     return nodes.map((node) => ({
@@ -300,7 +301,7 @@ export class SummaryIndexRetriever implements BaseRetriever {
 /**
  * LLM retriever for SummaryIndex which lets you select the most relevant chunks.
  */
-export class SummaryIndexLLMRetriever implements BaseRetriever {
+export class SummaryIndexLLMRetriever extends BaseRetriever {
   index: SummaryIndex;
   choiceSelectPrompt: ChoiceSelectPrompt;
   choiceBatchSize: number;
@@ -317,6 +318,7 @@ export class SummaryIndexLLMRetriever implements BaseRetriever {
     parseChoiceSelectAnswerFn?: ChoiceSelectParserFunction,
     serviceContext?: ServiceContext,
   ) {
+    super();
     this.index = index;
     this.choiceSelectPrompt = choiceSelectPrompt || defaultChoiceSelectPrompt;
     this.choiceBatchSize = choiceBatchSize;
@@ -326,7 +328,7 @@ export class SummaryIndexLLMRetriever implements BaseRetriever {
     this.serviceContext = serviceContext || index.serviceContext;
   }
 
-  async retrieve({ query }: RetrieveParams): Promise<NodeWithScore[]> {
+  async _retrieve(query: QueryBundle): Promise<NodeWithScore[]> {
     const nodeIds = this.index.indexStruct.nodes;
     const results: NodeWithScore[] = [];
 

packages/llamaindex/src/indices/vectorStore/index.ts

@@ -2,9 +2,10 @@ import {
   DEFAULT_SIMILARITY_TOP_K,
   type BaseEmbedding,
 } from "@llamaindex/core/embeddings";
-import { Settings } from "@llamaindex/core/global";
 import type { MessageContent } from "@llamaindex/core/llms";
+import type { QueryBundle } from "@llamaindex/core/query-engine";
 import type { BaseSynthesizer } from "@llamaindex/core/response-synthesizers";
+import { BaseRetriever } from "@llamaindex/core/retriever";
 import {
   ImageNode,
   ModalityType,
@@ -14,8 +15,6 @@ import {
   type Document,
   type NodeWithScore,
 } from "@llamaindex/core/schema";
-import { wrapEventCaller } from "@llamaindex/core/utils";
-import type { BaseRetriever, RetrieveParams } from "../../Retriever.js";
 import type { ServiceContext } from "../../ServiceContext.js";
 import { nodeParserFromSettingsOrContext } from "../../Settings.js";
 import { RetrieverQueryEngine } from "../../engines/query/RetrieverQueryEngine.js";
@@ -388,7 +387,7 @@ export type VectorIndexRetrieverOptions = {
   filters?: MetadataFilters;
 };
 
-export class VectorIndexRetriever implements BaseRetriever {
+export class VectorIndexRetriever extends BaseRetriever {
   index: VectorStoreIndex;
   topK: TopKMap;
 
@@ -401,6 +400,7 @@ export class VectorIndexRetriever implements BaseRetriever {
     topK,
     filters,
   }: VectorIndexRetrieverOptions) {
+    super();
     this.index = index;
     this.serviceContext = this.index.serviceContext;
     this.topK = topK ?? {
@@ -417,32 +417,17 @@ export class VectorIndexRetriever implements BaseRetriever {
     this.topK[ModalityType.TEXT] = similarityTopK;
   }
 
-  @wrapEventCaller
-  async retrieve({
-    query,
-    preFilters,
-  }: RetrieveParams): Promise<NodeWithScore[]> {
-    Settings.callbackManager.dispatchEvent("retrieve-start", {
-      query,
-    });
+  async _retrieve(params: QueryBundle): Promise<NodeWithScore[]> {
+    const { query } = params;
     const vectorStores = this.index.vectorStores;
     let nodesWithScores: NodeWithScore[] = [];
 
     for (const type in vectorStores) {
       const vectorStore: VectorStore = vectorStores[type as ModalityType]!;
       nodesWithScores = nodesWithScores.concat(
-        await this.retrieveQuery(
-          query,
-          type as ModalityType,
-          vectorStore,
-          preFilters as MetadataFilters,
-        ),
+        await this.retrieveQuery(query, type as ModalityType, vectorStore),
       );
     }
-    Settings.callbackManager.dispatchEvent("retrieve-end", {
-      query,
-      nodes: nodesWithScores,
-    });
     return nodesWithScores;
   }
 

packages/llamaindex/src/ingestion/IngestionCache.ts

@@ -1,7 +1,11 @@
 import type { BaseNode, TransformComponent } from "@llamaindex/core/schema";
 import { MetadataMode } from "@llamaindex/core/schema";
 import { createSHA256 } from "@llamaindex/env";
-import { docToJson, jsonToDoc } from "../storage/docStore/utils.js";
+import {
+  docToJson,
+  jsonSerializer,
+  jsonToDoc,
+} from "../storage/docStore/utils.js";
 import { SimpleKVStore } from "../storage/kvStore/SimpleKVStore.js";
 import type { BaseKVStore } from "../storage/kvStore/types.js";
 
@@ -53,7 +57,7 @@ export class IngestionCache {
 
   async put(hash: string, nodes: BaseNode[]) {
     const val = {
-      [this.nodesKey]: nodes.map((node) => docToJson(node)),
+      [this.nodesKey]: nodes.map((node) => docToJson(node, jsonSerializer)),
     };
     await this.cache.put(hash, val, this.collection);
   }
@@ -63,6 +67,8 @@ export class IngestionCache {
     if (!json || !json[this.nodesKey] || !Array.isArray(json[this.nodesKey])) {
       return undefined;
     }
-    return json[this.nodesKey].map((doc: any) => jsonToDoc(doc));
+    return json[this.nodesKey].map((doc: any) =>
+      jsonToDoc(doc, jsonSerializer),
+    );
   }
 }

packages/llamaindex/src/llm/types.ts

@@ -1,10 +0,0 @@
-import type { MessageContent } from "@llamaindex/core/llms";
-import type { NodeWithScore } from "@llamaindex/core/schema";
-
-export type RetrievalStartEvent = {
-  query: MessageContent;
-};
-export type RetrievalEndEvent = {
-  query: MessageContent;
-  nodes: NodeWithScore[];
-};

packages/llamaindex/src/objects/base.ts

@@ -1,8 +1,8 @@
 import type { BaseTool, MessageContent } from "@llamaindex/core/llms";
+import { BaseRetriever } from "@llamaindex/core/retriever";
 import type { BaseNode, Metadata } from "@llamaindex/core/schema";
 import { TextNode } from "@llamaindex/core/schema";
 import { extractText } from "@llamaindex/core/utils";
-import type { BaseRetriever } from "../Retriever.js";
 import type { VectorStoreIndex } from "../indices/vectorStore/index.js";
 
 // Assuming that necessary interfaces and classes (like OT, TextNode, BaseNode, etc.) are defined elsewhere
@@ -49,9 +49,6 @@ export abstract class BaseObjectNodeMapping {
 
 // You will need to implement specific subclasses of BaseObjectNodeMapping as per your project requirements.
 
-// todo: multimodal support
-type QueryType = MessageContent;
-
 export class ObjectRetriever<T = unknown> {
   _retriever: BaseRetriever;
   _objectNodeMapping: BaseObjectNodeMapping;
@@ -70,7 +67,7 @@ export class ObjectRetriever<T = unknown> {
   }
 
   // Translating the retrieve method
-  async retrieve(strOrQueryBundle: QueryType): Promise<T[]> {
+  async retrieve(strOrQueryBundle: MessageContent): Promise<T[]> {
     const nodes = await this.retriever.retrieve({
       query: extractText(strOrQueryBundle),
     });

packages/llamaindex/src/storage/docStore/KVDocumentStore.ts

@@ -29,7 +29,7 @@ export class KVDocumentStore extends BaseDocumentStore {
     for (const key in jsonDict) {
       const value = jsonDict[key];
       if (isValidDocJson(value)) {
-        docs[key] = jsonToDoc(value);
+        docs[key] = jsonToDoc(value, this.serializer);
       } else {
         console.warn(`Invalid JSON for docId ${key}`);
       }
@@ -52,7 +52,7 @@ export class KVDocumentStore extends BaseDocumentStore {
         );
       }
       const nodeKey = doc.id_;
-      const data = docToJson(doc);
+      const data = docToJson(doc, this.serializer);
       await this.kvstore.put(nodeKey, data, this.nodeCollection);
       const metadata: DocMetaData = { docHash: doc.hash };
 
@@ -94,7 +94,7 @@ export class KVDocumentStore extends BaseDocumentStore {
     if (!isValidDocJson(json)) {
       throw new Error(`Invalid JSON for docId ${docId}`);
     }
-    return jsonToDoc(json);
+    return jsonToDoc(json, this.serializer);
   }
 
   async getRefDocInfo(refDocId: string): Promise<RefDocInfo | undefined> {

packages/llamaindex/src/storage/docStore/PostgresDocumentStore.ts

@@ -4,6 +4,7 @@ import {
   type PostgresKVStoreConfig,
 } from "../kvStore/PostgresKVStore.js";
 import { KVDocumentStore } from "./KVDocumentStore.js";
+import { noneSerializer } from "./utils.js";
 
 const DEFAULT_TABLE_NAME = "llamaindex_doc_store";
 
@@ -12,6 +13,8 @@ export type PostgresDocumentStoreConfig = PostgresKVStoreConfig & {
 };
 
 export class PostgresDocumentStore extends KVDocumentStore {
+  serializer = noneSerializer;
+
   constructor(config?: PostgresDocumentStoreConfig) {
     const kvStore = new PostgresKVStore({
       schemaName: config?.schemaName,

packages/llamaindex/src/storage/docStore/types.ts

@@ -3,6 +3,7 @@ import {
   DEFAULT_PERSIST_DIR,
 } from "@llamaindex/core/global";
 import { BaseNode } from "@llamaindex/core/schema";
+import { jsonSerializer, type Serializer } from "./utils.js";
 
 const defaultPersistPath = `${DEFAULT_PERSIST_DIR}/${DEFAULT_DOC_STORE_PERSIST_FILENAME}`;
 
@@ -12,6 +13,8 @@ export interface RefDocInfo {
 }
 
 export abstract class BaseDocumentStore {
+  serializer: Serializer<any> = jsonSerializer;
+
   // Save/load
   persist(persistPath: string = defaultPersistPath): void {
     // Persist the docstore to a file.

packages/llamaindex/src/storage/docStore/utils.ts

@@ -4,12 +4,35 @@ import { Document, ObjectType, TextNode } from "@llamaindex/core/schema";
 const TYPE_KEY = "__type__";
 const DATA_KEY = "__data__";
 
-type DocJson = {
-  [TYPE_KEY]: ObjectType;
-  [DATA_KEY]: string;
+export interface Serializer<T> {
+  toPersistence(data: Record<string, unknown>): T;
+  fromPersistence(data: T): Record<string, unknown>;
+}
+
+export const jsonSerializer: Serializer<string> = {
+  toPersistence(data) {
+    return JSON.stringify(data);
+  },
+  fromPersistence(data) {
+    return JSON.parse(data);
+  },
 };
 
-export function isValidDocJson(docJson: any): docJson is DocJson {
+export const noneSerializer: Serializer<Record<string, unknown>> = {
+  toPersistence(data) {
+    return data;
+  },
+  fromPersistence(data) {
+    return data;
+  },
+};
+
+type DocJson<Data> = {
+  [TYPE_KEY]: ObjectType;
+  [DATA_KEY]: Data;
+};
+
+export function isValidDocJson(docJson: any): docJson is DocJson<unknown> {
   return (
     typeof docJson === "object" &&
     docJson !== null &&
@@ -18,16 +41,22 @@ export function isValidDocJson(docJson: any): docJson is DocJson {
   );
 }
 
-export function docToJson(doc: BaseNode): DocJson {
+export function docToJson(
+  doc: BaseNode,
+  serializer: Serializer<unknown>,
+): DocJson<unknown> {
   return {
-    [DATA_KEY]: JSON.stringify(doc.toJSON()),
+    [DATA_KEY]: serializer.toPersistence(doc.toJSON()),
     [TYPE_KEY]: doc.type,
   };
 }
 
-export function jsonToDoc(docDict: DocJson): BaseNode {
+export function jsonToDoc<Data>(
+  docDict: DocJson<Data>,
+  serializer: Serializer<Data>,
+): BaseNode {
   const docType = docDict[TYPE_KEY];
-  const dataDict = JSON.parse(docDict[DATA_KEY]);
+  const dataDict = serializer.fromPersistence(docDict[DATA_KEY]) as any;
   let doc: BaseNode;
 
   if (docType === ObjectType.DOCUMENT) {

packages/llm/groq/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @llamaindex/groq
 
+## 0.0.7
+
+### Patch Changes
+
+- @llamaindex/openai@0.1.8
+
 ## 0.0.6
 
 ### Patch Changes

packages/llm/groq/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/groq",
   "description": "Groq Adapter for LlamaIndex",
-  "version": "0.0.6",
+  "version": "0.0.7",
   "type": "module",
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",

packages/llm/openai/CHANGELOG.md

@@ -1,5 +1,12 @@
 # @llamaindex/openai
 
+## 0.1.8
+
+### Patch Changes
+
+- Updated dependencies [8b7fdba]
+  - @llamaindex/core@0.2.6
+
 ## 0.1.7
 
 ### Patch Changes

packages/llm/openai/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/openai",
   "description": "OpenAI Adapter for LlamaIndex",
-  "version": "0.1.7",
+  "version": "0.1.8",
   "type": "module",
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",