Release 0.5.2 (#1035)

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

.github/workflows/lint_on_push_or_pull.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

.github/workflows/preview.yml

@@ -10,7 +10,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/publish_github.yml

@@ -12,7 +12,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/release.yml

@@ -15,7 +15,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4
@@ -67,3 +67,4 @@ jobs:
         with:
           commit_message: "chore: update lock file"
           branch: changeset-release/main
+          file_pattern: "pnpm-lock.yaml"

.github/workflows/test.yml

@@ -23,7 +23,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4
@@ -45,7 +45,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -60,7 +60,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -97,7 +97,7 @@ jobs:
     name: Build LlamaIndex Example (${{ matrix.packages }})
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -116,7 +116,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

apps/docs/CHANGELOG.md

@@ -1,5 +1,52 @@
 # docs
 
+## 0.0.43
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.0.42
+
+### Patch Changes
+
+- 2774681: Add mixedbread's embeddings and reranking API
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.0.41
+
+### Patch Changes
+
+- 36ddec4: fix: typo in custom page separator parameter for LlamaParse
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+  - @llamaindex/examples@0.0.7
+
+## 0.0.40
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.39
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.38
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.0.37
 
 ### Patch Changes

apps/docs/docs/getting_started/concepts.md

@@ -62,7 +62,7 @@ These building blocks can be customized to reflect ranking preferences, as well
 
 [**Retrievers**](../modules/retriever.md):
 A retriever defines how to efficiently retrieve relevant context from a knowledge base (i.e. index) when given a query.
-The specific retrieval logic differs for difference indices, the most popular being dense retrieval against a vector index.
+The specific retrieval logic differs for different indices, the most popular being dense retrieval against a vector index.
 
 [**Response Synthesizers**](../modules/response_synthesizer.md):
 A response synthesizer generates a response from an LLM, using a user query and a given set of retrieved text chunks.

apps/docs/docs/modules/data_loaders/llama_parse/index.mdx

@@ -27,23 +27,25 @@ They can be divided into two groups.
 
 - `apiKey` is required. Can be set as an environment variable `LLAMA_CLOUD_API_KEY`
 - `checkInterval` is the interval in seconds to check if the parsing is done. Default is `1`.
-- `maxTimeout` is the maximum timout to wait for parsing to finish. Default is `2000`
+- `maxTimeout` is the maximum timeout to wait for parsing to finish. Default is `2000`
 - `verbose` shows progress of the parsing. Default is `true`
 - `ignoreErrors` set to false to get errors while parsing. Default is `true` and returns an empty array on error.
 
 #### Advanced params:
 
 - `resultType` can be set to `markdown`, `text` or `json`. Defaults to `text`. More information about `json` mode on the next pages.
-- `language` primarly helps with OCR recognition. Defaults to `en`. Click [here](../../../api/type-aliases/Language.md) for a list of supported languages.
+- `language` primarily helps with OCR recognition. Defaults to `en`. Click [here](../../../api/type-aliases/Language.md) for a list of supported languages.
 - `parsingInstructions?` Optional. Can help with complicated document structures. See this [LlamaIndex Blog Post](https://www.llamaindex.ai/blog/launching-the-first-genai-native-document-parsing-platform) for an example.
 - `skipDiagonalText?` Optional. Set to true to ignore diagonal text. (Text that is not rotated 0, 90, 180 or 270 degrees)
 - `invalidateCache?` Optional. Set to true to ignore the LlamaCloud cache. All document are kept in cache for 48hours after the job was completed to avoid processing the same document twice. Can be useful for testing when trying to re-parse the same document with, e.g. different `parsingInstructions`.
 - `doNotCache?` Optional. Set to true to not cache the document.
 - `fastMode?` Optional. Set to true to use the fast mode. This mode will skip OCR of images, and table/heading reconstruction. Note: Non-compatible with `gpt4oMode`.
-- `doNotUnrollColumns?` Optional. Set to true to keep the text according to document layout. Reduce reconstruction accuracy, and LLM's/embedings performances in most cases.
-- `pageSeperator?` Optional. The page seperator to use. Defaults is `\\n---\\n`.
+- `doNotUnrollColumns?` Optional. Set to true to keep the text according to document layout. Reduce reconstruction accuracy, and LLMs/embeddings performances in most cases.
+- `pageSeparator?` Optional. The page separator to use. Defaults is `\\n---\\n`.
 - `gpt4oMode` set to true to use GPT-4o to extract content. Default is `false`.
 - `gpt4oApiKey?` Optional. Set the GPT-4o API key. Lowers the cost of parsing by using your own API key. Your OpenAI account will be charged. Can also be set in the environment variable `LLAMA_CLOUD_GPT4O_API_KEY`.
+- `boundingBox?` Optional. Specify an area of the document to parse. Expects the bounding box margins as a string in clockwise order, e.g. `boundingBox = "0.1,0,0,0"` to not parse the top 10% of the document.
+- `targetPages?` Optional. Specify which pages to parse by specifying them as a comma-separated list. First page is `0`.
 - `numWorkers` as in the python version, is set in `SimpleDirectoryReader`. Default is 1.
 
 ### LlamaParse with SimpleDirectoryReader

apps/docs/docs/modules/data_loaders/llama_parse/json_mode.mdx

@@ -8,7 +8,7 @@ In JSON mode, LlamaParse will return a data structure representing the parsed ob
 
 ## Usage
 
-For Json mode, you need to use `loadJson`. The `resultType` is automatically set with this method. Currently it can't be used with `SimpleDirectoryReader`.
+For Json mode, you need to use `loadJson`. The `resultType` is automatically set with this method.
 More information about indexing the results on the next page.
 
 ```ts

apps/docs/docs/modules/embeddings/available_embeddings/mixedbreadai.md

@@ -0,0 +1,100 @@
+# MixedbreadAI
+
+Welcome to the mixedbread embeddings guide! This guide will help you use the mixedbread ai's API to generate embeddings for your text documents, ensuring you get the most relevant information, just like picking the freshest bread from the bakery.
+
+To find out more about the latest features, updates, and available models, visit [mixedbread.ai](https://mixedbread-ai.com/).
+
+## Table of Contents
+
+1. [Setup](#setup)
+2. [Usage with LlamaIndex](#integration-with-llamaindex)
+3. [Embeddings with Custom Parameters](#embeddings-with-custom-parameters)
+
+## Setup
+
+First, you will need to install the `llamaindex` package.
+
+```bash
+pnpm install llamaindex
+```
+
+Next, sign up for an API key at [mixedbread.ai](https://mixedbread.ai/). Once you have your API key, you can import the necessary modules and create a new instance of the `MixedbreadAIEmbeddings` class.
+
+```ts
+import { MixedbreadAIEmbeddings, Document, Settings } from "llamaindex";
+```
+
+## Usage with LlamaIndex
+
+This section will guide you through integrating mixedbread embeddings with LlamaIndex for more advanced usage.
+
+### Step 1: Load and Index Documents
+
+For this example, we will use a single document. In a real-world scenario, you would have multiple documents to index, like a variety of breads in a bakery.
+
+```ts
+Settings.embedModel = new MixedbreadAIEmbeddings({
+  apiKey: "<MIXEDBREAD_API_KEY>",
+  model: "mixedbread-ai/mxbai-embed-large-v1",
+});
+
+const document = new Document({
+  text: "The true source of happiness.",
+  id_: "bread",
+});
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+```
+
+### Step 2: Create a Query Engine
+
+Combine the retriever and the embed model to create a query engine. This setup ensures that your queries are processed to provide the best results, like arranging the bread in the order of freshness and quality.
+
+Models can require prompts to generate embeddings for queries, in the 'mixedbread-ai/mxbai-embed-large-v1' model's case, the prompt is `Represent this sentence for searching relevant passages:`.
+
+```ts
+const queryEngine = index.asQueryEngine();
+
+const query =
+  "Represent this sentence for searching relevant passages: What is bread?";
+
+// Log the response
+const results = await queryEngine.query(query);
+console.log(results); // Serving up the freshest, most relevant results.
+```
+
+## Embeddings with Custom Parameters
+
+This section will guide you through generating embeddings with custom parameters and usage with f.e. matryoshka and binary embeddings.
+
+### Step 1: Create an Instance of MixedbreadAIEmbeddings
+
+Create a new instance of the `MixedbreadAIEmbeddings` class with custom parameters. For example, to use the `mixedbread-ai/mxbai-embed-large-v1` model with a batch size of 64, normalized embeddings, and binary encoding format:
+
+```ts
+const embeddings = new MixedbreadAIEmbeddings({
+  apiKey: "<MIXEDBREAD_API_KEY>",
+  model: "mixedbread-ai/mxbai-embed-large-v1",
+  batchSize: 64,
+  normalized: true,
+  dimensions: 512,
+  encodingFormat: MixedbreadAI.EncodingFormat.Binary,
+});
+```
+
+### Step 2: Define Texts
+
+Define the texts you want to generate embeddings for.
+
+```ts
+const texts = ["Bread is life", "Bread is love"];
+```
+
+### Step 3: Generate Embeddings
+
+Use the `embedDocuments` method to generate embeddings for the texts.
+
+```ts
+const result = await embeddings.embedDocuments(texts);
+console.log(result); // Perfectly customized embeddings, ready to serve.
+```

apps/docs/docs/modules/llms/available_llms/bedrock.md

@@ -15,7 +15,7 @@ Settings.llm = new Bedrock({
 });
 ```
 
-Currently only supports Anthropic models:
+Currently only supports Anthropic and Meta models:
 
 ```ts
 ANTHROPIC_CLAUDE_INSTANT_1 = "anthropic.claude-instant-v1";
@@ -25,6 +25,10 @@ ANTHROPIC_CLAUDE_3_SONNET = "anthropic.claude-3-sonnet-20240229-v1:0";
 ANTHROPIC_CLAUDE_3_HAIKU = "anthropic.claude-3-haiku-20240307-v1:0";
 ANTHROPIC_CLAUDE_3_OPUS = "anthropic.claude-3-opus-20240229-v1:0"; // available on us-west-2
 ANTHROPIC_CLAUDE_3_5_SONNET = "anthropic.claude-3-5-sonnet-20240620-v1:0";
+META_LLAMA2_13B_CHAT = "meta.llama2-13b-chat-v1";
+META_LLAMA2_70B_CHAT = "meta.llama2-70b-chat-v1";
+META_LLAMA3_8B_INSTRUCT = "meta.llama3-8b-instruct-v1:0";
+META_LLAMA3_70B_INSTRUCT = "meta.llama3-70b-instruct-v1:0";
 ```
 
 Sonnet, Haiku and Opus are multimodal, image_url only supports base64 data url format, e.g. `data:image/jpeg;base64,SGVsbG8sIFdvcmxkIQ==`

apps/docs/docs/modules/node_postprocessors/mixedbreadiai_reranker.md

@@ -0,0 +1,164 @@
+# MixedbreadAI
+
+Welcome to the mixedbread ai reranker guide! This guide will help you use mixedbread ai's API to rerank search query results, ensuring you get the most relevant information, just like picking the freshest bread from the bakery.
+
+To find out more about the latest features and updates, visit the [mixedbread.ai](https://mixedbread.ai/).
+
+## Table of Contents
+
+1. [Setup](#setup)
+2. [Usage with LlamaIndex](#integration-with-llamaindex)
+3. [Simple Reranking Guide](#simple-reranking-guide)
+4. [Reranking with Objects](#reranking-with-objects)
+
+## Setup
+
+First, you will need to install the `llamaindex` package.
+
+```bash
+pnpm install llamaindex
+```
+
+Next, sign up for an API key at [mixedbread.ai](https://mixedbread.ai/). Once you have your API key, you can import the necessary modules and create a new instance of the `MixedbreadAIReranker` class.
+
+```ts
+import {
+  MixedbreadAIReranker,
+  Document,
+  OpenAI,
+  VectorStoreIndex,
+  Settings,
+} from "llamaindex";
+```
+
+## Usage with LlamaIndex
+
+This section will guide you through integrating mixedbread's reranker with LlamaIndex.
+
+### Step 1: Load and Index Documents
+
+For this example, we will use a single document. In a real-world scenario, you would have multiple documents to index, like a variety of breads in a bakery.
+
+```ts
+const document = new Document({
+  text: "This is a sample document.",
+  id_: "sampleDoc",
+});
+
+Settings.llm = new OpenAI({ model: "gpt-3.5-turbo", temperature: 0.1 });
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+```
+
+### Step 2: Increase Similarity TopK
+
+The default value for `similarityTopK` is 2, which means only the most similar document will be returned. To get more results, like picking a variety of fresh breads, you can increase the value of `similarityTopK`.
+
+```ts
+const retriever = index.asRetriever();
+retriever.similarityTopK = 5;
+```
+
+### Step 3: Create a MixedbreadAIReranker Instance
+
+Create a new instance of the `MixedbreadAIReranker` class.
+
+```ts
+const nodePostprocessor = new MixedbreadAIReranker({
+  apiKey: "<MIXEDBREAD_API_KEY>",
+  topN: 4,
+});
+```
+
+### Step 4: Create a Query Engine
+
+Combine the retriever and node postprocessor to create a query engine. This setup ensures that your queries are processed and reranked to provide the best results, like arranging the bread in the order of freshness and quality.
+
+```ts
+const queryEngine = index.asQueryEngine({
+  retriever,
+  nodePostprocessors: [nodePostprocessor],
+});
+
+// Log the response
+const response = await queryEngine.query("Where did the author grow up?");
+console.log(response);
+```
+
+With mixedbread's Reranker, you're all set to serve up the most relevant and well-ordered results, just like a skilled baker arranging their best breads for eager customers. Enjoy the perfect blend of technology and culinary delight!
+
+## Simple Reranking Guide
+
+This section will guide you through a simple reranking process using mixedbread ai.
+
+### Step 1: Create an Instance of MixedbreadAIReranker
+
+Create a new instance of the `MixedbreadAIReranker` class, passing in your API key and the number of results you want to return. It's like setting up your bakery to offer a specific number of freshly baked items.
+
+```ts
+const reranker = new MixedbreadAIReranker({
+  apiKey: "<MIXEDBREAD_API_KEY>",
+  topN: 4,
+});
+```
+
+### Step 2: Define Nodes and Query
+
+Define the nodes (documents) you want to rerank and the query.
+
+```ts
+const nodes = [
+  { node: new BaseNode("To bake bread you need flour") },
+  { node: new BaseNode("To bake bread you need yeast") },
+];
+const query = "What do you need to bake bread?";
+```
+
+### Step 3: Perform Reranking
+
+Use the `postprocessNodes` method to rerank the nodes based on the query.
+
+```ts
+const result = await reranker.postprocessNodes(nodes, query);
+console.log(result); // Like pulling freshly baked nodes out of the oven.
+```
+
+## Reranking with Objects
+
+This section will guide you through reranking when working with objects.
+
+### Step 1: Create an Instance of MixedbreadAIReranker
+
+Create a new instance of the `MixedbreadAIReranker` class, just like before.
+
+```ts
+const reranker = new MixedbreadAIReranker({
+  apiKey: "<MIXEDBREAD_API_KEY>",
+  model: "mixedbread-ai/mxbai-rerank-large-v1",
+  topK: 5,
+  rankFields: ["title", "content"],
+  returnInput: true,
+  maxRetries: 5,
+});
+```
+
+### Step 2: Define Documents and Query
+
+Define the documents (objects) you want to rerank and the query.
+
+```ts
+const documents = [
+  { title: "Bread Recipe", content: "To bake bread you need flour" },
+  { title: "Bread Recipe", content: "To bake bread you need yeast" },
+];
+const query = "What do you need to bake bread?";
+```
+
+### Step 3: Perform Reranking
+
+Use the `rerank` method to reorder the documents based on the query.
+
+```ts
+const result = await reranker.rerank(documents, query);
+console.log(result); // Perfectly customized results, ready to serve.
+```

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docs",
-  "version": "0.0.37",
+  "version": "0.0.43",
   "private": true,
   "scripts": {
     "docusaurus": "docusaurus",
@@ -15,28 +15,28 @@
     "typecheck": "tsc"
   },
   "dependencies": {
-    "@docusaurus/core": "^3.3.2",
-    "@docusaurus/remark-plugin-npm2yarn": "^3.3.2",
+    "@docusaurus/core": "3.4.0",
+    "@docusaurus/remark-plugin-npm2yarn": "3.4.0",
     "@llamaindex/examples": "workspace:*",
-    "@mdx-js/react": "^3.0.1",
-    "clsx": "^2.1.1",
+    "@mdx-js/react": "3.0.1",
+    "clsx": "2.1.1",
     "llamaindex": "workspace:*",
-    "postcss": "^8.4.38",
-    "prism-react-renderer": "^2.3.1",
-    "raw-loader": "^4.0.2",
-    "react": "^18.3.1",
-    "react-dom": "^18.3.1"
+    "postcss": "8.4.39",
+    "prism-react-renderer": "2.3.1",
+    "raw-loader": "4.0.2",
+    "react": "18.3.1",
+    "react-dom": "18.3.1"
   },
   "devDependencies": {
-    "@docusaurus/module-type-aliases": "3.3.2",
-    "@docusaurus/preset-classic": "^3.3.2",
-    "@docusaurus/theme-classic": "^3.3.2",
-    "@docusaurus/types": "^3.3.2",
-    "@tsconfig/docusaurus": "^2.0.3",
+    "@docusaurus/module-type-aliases": "3.4.0",
+    "@docusaurus/preset-classic": "3.4.0",
+    "@docusaurus/theme-classic": "3.4.0",
+    "@docusaurus/types": "3.4.0",
+    "@tsconfig/docusaurus": "2.0.3",
     "@types/node": "^20.12.11",
-    "docusaurus-plugin-typedoc": "^1.0.1",
-    "typedoc": "^0.25.13",
-    "typedoc-plugin-markdown": "^4.0.1",
+    "docusaurus-plugin-typedoc": "1.0.2",
+    "typedoc": "0.26.3",
+    "typedoc-plugin-markdown": "4.1.2",
     "typescript": "^5.5.2"
   },
   "browserslist": {

examples/CHANGELOG.md

@@ -1,5 +1,15 @@
 # examples
 
+## 0.0.7
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+  - @llamaindex/core@0.1.0
+
 ## 0.0.6
 
 ### Patch Changes

examples/anthropic/agent.ts

@@ -2,7 +2,7 @@ import { Anthropic, FunctionTool, Settings, WikipediaTool } from "llamaindex";
 import { AnthropicAgent } from "llamaindex/agent/anthropic";
 
 Settings.callbackManager.on("llm-tool-call", (event) => {
-  console.log("llm-tool-call", event.detail.payload.toolCall);
+  console.log("llm-tool-call", event.detail.toolCall);
 });
 
 const anthropic = new Anthropic({

examples/multimodal/context.ts

@@ -4,7 +4,6 @@ import {
   NodeWithScore,
   ObjectType,
   OpenAI,
-  RetrievalEndEvent,
   Settings,
   VectorStoreIndex,
 } from "llamaindex";
@@ -18,8 +17,8 @@ Settings.chunkOverlap = 20;
 Settings.llm = new OpenAI({ model: "gpt-4-turbo", maxTokens: 512 });
 
 // Update callbackManager
-Settings.callbackManager.on("retrieve-end", (event: RetrievalEndEvent) => {
-  const { nodes, query } = event.detail.payload;
+Settings.callbackManager.on("retrieve-end", (event) => {
+  const { nodes, query } = event.detail;
   const imageNodes = nodes.filter(
     (node: NodeWithScore) => node.node.type === ObjectType.IMAGE_DOCUMENT,
   );

examples/multimodal/rag.ts

@@ -1,7 +1,6 @@
 import {
   MultiModalResponseSynthesizer,
   OpenAI,
-  RetrievalEndEvent,
   Settings,
   VectorStoreIndex,
 } from "llamaindex";
@@ -15,8 +14,8 @@ Settings.chunkOverlap = 20;
 Settings.llm = new OpenAI({ model: "gpt-4-turbo", maxTokens: 512 });
 
 // Update callbackManager
-Settings.callbackManager.on("retrieve-end", (event: RetrievalEndEvent) => {
-  const { nodes, query } = event.detail.payload;
+Settings.callbackManager.on("retrieve-end", (event) => {
+  const { nodes, query } = event.detail;
   console.log(`Retrieved ${nodes.length} nodes for query: ${query}`);
 });
 

examples/package.json

@@ -1,11 +1,12 @@
 {
   "name": "@llamaindex/examples",
   "private": true,
-  "version": "0.0.6",
+  "version": "0.0.7",
   "dependencies": {
     "@aws-crypto/sha256-js": "^5.2.0",
     "@azure/identity": "^4.2.1",
     "@datastax/astra-db-ts": "^1.2.1",
+    "@llamaindex/core": "^0.1.0",
     "@notionhq/client": "^2.2.15",
     "@pinecone-database/pinecone": "^2.2.2",
     "@zilliz/milvus2-sdk-node": "^2.4.2",
@@ -13,7 +14,7 @@
     "commander": "^12.1.0",
     "dotenv": "^16.4.5",
     "js-tiktoken": "^1.0.12",
-    "llamaindex": "^0.4.3",
+    "llamaindex": "^0.5.0",
     "mongodb": "^6.7.0",
     "pathe": "^1.1.2"
   },

examples/qdrantdb/preFilters.ts

@@ -1,8 +1,8 @@
 import * as dotenv from "dotenv";
 import {
-  CallbackManager,
   Document,
   MetadataMode,
+  NodeWithScore,
   QdrantVectorStore,
   Settings,
   VectorStoreIndex,
@@ -10,13 +10,12 @@ import {
 } from "llamaindex";
 
 // Update callback manager
-Settings.callbackManager = new CallbackManager({
-  onRetrieve: (data) => {
-    console.log(
-      "The retrieved nodes are:",
-      data.nodes.map((node) => node.node.getContent(MetadataMode.NONE)),
-    );
-  },
+Settings.callbackManager.on("retrieve-end", (event) => {
+  const { nodes } = event.detail;
+  console.log(
+    "The retrieved nodes are:",
+    nodes.map((node: NodeWithScore) => node.node.getContent(MetadataMode.NONE)),
+  );
 });
 
 // Load environment variables from local .env file

examples/recipes/cost-analysis.ts

@@ -1,7 +1,7 @@
+import { extractText } from "@llamaindex/core/utils";
 import { encodingForModel } from "js-tiktoken";
-import { ChatMessage, OpenAI, type LLMStartEvent } from "llamaindex";
+import { ChatMessage, OpenAI } from "llamaindex";
 import { Settings } from "llamaindex/Settings";
-import { extractText } from "llamaindex/llm/utils";
 
 const encoding = encodingForModel("gpt-4-0125-preview");
 
@@ -12,8 +12,8 @@ const llm = new OpenAI({
 
 let tokenCount = 0;
 
-Settings.callbackManager.on("llm-start", (event: LLMStartEvent) => {
-  const { messages } = event.detail.payload;
+Settings.callbackManager.on("llm-start", (event) => {
+  const { messages } = event.detail;
   messages.reduce((count: number, message: ChatMessage) => {
     return count + encoding.encode(extractText(message.content)).length;
   }, 0);
@@ -24,7 +24,7 @@ Settings.callbackManager.on("llm-start", (event: LLMStartEvent) => {
 });
 
 Settings.callbackManager.on("llm-stream", (event) => {
-  const { chunk } = event.detail.payload;
+  const { chunk } = event.detail;
   const { delta } = chunk;
   tokenCount += encoding.encode(extractText(delta)).length;
   if (tokenCount > 20) {

package.json

@@ -39,9 +39,6 @@
       "trim": "1.0.1",
       "@babel/traverse": "7.23.2",
       "protobufjs": "7.2.6"
-    },
-    "patchedDependencies": {
-      "bunchee@5.2.1": "patches/bunchee@5.2.1.patch"
     }
   },
   "lint-staged": {

packages/autotool/CHANGELOG.md

@@ -1,5 +1,14 @@
 # @llamaindex/autotool
 
+## 2.0.0
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
 ## 1.0.0
 
 ### Patch Changes

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

@@ -1,5 +1,55 @@
 # @llamaindex/autotool-02-next-example
 
+## 0.1.27
+
+### Patch Changes
+
+- llamaindex@0.5.2
+- @llamaindex/autotool@2.0.0
+
+## 0.1.26
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+  - @llamaindex/autotool@2.0.0
+
+## 0.1.25
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+  - @llamaindex/autotool@2.0.0
+
+## 0.1.24
+
+### Patch Changes
+
+- llamaindex@0.4.14
+- @llamaindex/autotool@1.0.0
+
+## 0.1.23
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.22
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+  - @llamaindex/autotool@1.0.0
+
 ## 0.1.21
 
 ### Patch Changes

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

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool-02-next-example",
   "private": true,
-  "version": "0.1.21",
+  "version": "0.1.27",
   "scripts": {
     "dev": "next dev",
     "build": "next build",

packages/autotool/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool",
   "type": "module",
-  "version": "1.0.0",
+  "version": "2.0.0",
   "description": "auto transpile your JS function to LLM Agent compatible",
   "files": [
     "dist",
@@ -51,7 +51,7 @@
     "unplugin": "^1.10.1"
   },
   "peerDependencies": {
-    "llamaindex": "^0.4.11",
+    "llamaindex": "^0.5.2",
     "openai": "^4",
     "typescript": "^4"
   },
@@ -70,7 +70,7 @@
     "@swc/types": "^0.1.8",
     "@types/json-schema": "^7.0.15",
     "@types/node": "^20.12.11",
-    "bunchee": "^5.2.1",
+    "bunchee": "5.3.0-beta.0",
     "llamaindex": "workspace:*",
     "next": "14.2.3",
     "rollup": "^4.18.0",

packages/cloud/CHANGELOG.md

@@ -1,5 +1,29 @@
 # @llamaindex/cloud
 
+## 0.2.0
+
+### Minor Changes
+
+- 3ed6acc: feat: cloud api change
+
+## 0.1.4
+
+### Patch Changes
+
+- 36ddec4: fix: typo in custom page separator parameter for LlamaParse
+
+## 0.1.3
+
+### Patch Changes
+
+- 1c444d5: feat(cloud): update openapi.json
+
+## 0.1.2
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+
 ## 0.1.1
 
 ### Patch Changes

packages/cloud/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloud",
-  "version": "0.1.1",
+  "version": "0.2.0",
   "type": "module",
   "license": "MIT",
   "scripts": {
@@ -35,6 +35,6 @@
   },
   "devDependencies": {
     "@hey-api/openapi-ts": "^0.48.0",
-    "bunchee": "^5.2.1"
+    "bunchee": "5.3.0-beta.0"
   }
 }

packages/community/CHANGELOG.md

@@ -1,5 +1,42 @@
 # @llamaindex/community
 
+## 0.0.20
+
+### Patch Changes
+
+- 56746c2: fix: llama3 patched to handle empty content (can happen with system) and added max tokens export
+
+## 0.0.19
+
+### Patch Changes
+
+- 16ef5dd: refactor: depends on core pacakge instead of llamaindex
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+  - @llamaindex/core@0.1.0
+
+## 0.0.18
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.16
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.0.15
 
 ### Patch Changes

packages/community/README.md

@@ -5,6 +5,7 @@
 ## Current Features:
 
 - Bedrock support for the Anthropic Claude Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
+- Bedrock support for the Meta LLama 2 and 3 Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
 
 ## LICENSE
 

packages/community/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/community",
   "description": "Community package for LlamaIndexTS",
-  "version": "0.0.15",
+  "version": "0.0.20",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",
@@ -38,20 +38,15 @@
     "directory": "packages/community"
   },
   "scripts": {
-    "build": "rm -rf ./dist && pnpm run build:code && pnpm run build:type",
-    "build:code": "tsup",
-    "build:type": "tsc -p tsconfig.json",
-    "dev": "concurrently \"pnpm run build:esm --watch\" \"pnpm run build:cjs --watch\" \"pnpm run build:type --watch\""
+    "build": "bunchee",
+    "dev": "bunchee --watch"
   },
   "devDependencies": {
-    "@swc/cli": "^0.3.12",
-    "@swc/core": "^1.6.3",
-    "concurrently": "^8.2.2",
-    "tsup": "^8.1.0"
+    "@types/node": "^20.14.2",
+    "bunchee": "5.3.0-beta.0"
   },
   "dependencies": {
     "@aws-sdk/client-bedrock-runtime": "^3.600.0",
-    "@types/node": "^20.14.2",
-    "llamaindex": "workspace:*"
+    "@llamaindex/core": "workspace:*"
   }
 }

packages/community/src/index.ts

@@ -1 +1,5 @@
-export { BEDROCK_MODELS, Bedrock } from "./llm/bedrock/base.js";
+export {
+  BEDROCK_MODELS,
+  BEDROCK_MODEL_MAX_TOKENS,
+  Bedrock,
+} from "./llm/bedrock/base.js";

packages/community/src/llm/bedrock/base.ts

@@ -1,53 +1,35 @@
 import {
   BedrockRuntimeClient,
+  type BedrockRuntimeClientConfig,
   InvokeModelCommand,
   InvokeModelWithResponseStreamCommand,
-  ResponseStream,
-  type BedrockRuntimeClientConfig,
-  type InvokeModelCommandInput,
-  type InvokeModelWithResponseStreamCommandInput,
 } from "@aws-sdk/client-bedrock-runtime";
-import type {
-  BaseTool,
-  ChatMessage,
-  ChatResponse,
-  ChatResponseChunk,
-  CompletionResponse,
-  LLMChatParamsNonStreaming,
-  LLMChatParamsStreaming,
-  LLMCompletionParamsNonStreaming,
-  LLMCompletionParamsStreaming,
-  LLMMetadata,
-  PartialToolCall,
-  ToolCall,
-  ToolCallLLMMessageOptions,
-} from "llamaindex";
-import { ToolCallLLM, streamConverter, wrapLLMEvent } from "llamaindex";
-import type {
-  AnthropicNoneStreamingResponse,
-  AnthropicTextContent,
-  StreamEvent,
-  ToolBlock,
-  ToolChoice,
-} from "./types.js";
 import {
-  mapBaseToolsToAnthropicTools,
-  mapChatMessagesToAnthropicMessages,
-  mapMessageContentToMessageContentDetails,
-  toUtf8,
-} from "./utils.js";
-
-export type BedrockAdditionalChatOptions = { toolChoice: ToolChoice };
+  type ChatMessage,
+  type ChatResponse,
+  type CompletionResponse,
+  type LLMChatParamsNonStreaming,
+  type LLMChatParamsStreaming,
+  type LLMCompletionParamsNonStreaming,
+  type LLMCompletionParamsStreaming,
+  type LLMMetadata,
+  ToolCallLLM,
+  type ToolCallLLMMessageOptions,
+} from "@llamaindex/core/llms";
+import { streamConverter, wrapLLMEvent } from "@llamaindex/core/utils";
+import {
+  type BedrockAdditionalChatOptions,
+  type BedrockChatStreamResponse,
+  Provider,
+} from "./provider";
+import { PROVIDERS } from "./providers";
+import { mapMessageContentToMessageContentDetails } from "./utils.js";
 
 export type BedrockChatParamsStreaming = LLMChatParamsStreaming<
   BedrockAdditionalChatOptions,
   ToolCallLLMMessageOptions
 >;
 
-export type BedrockChatStreamResponse = AsyncIterable<
-  ChatResponseChunk<ToolCallLLMMessageOptions>
->;
-
 export type BedrockChatParamsNonStreaming = LLMChatParamsNonStreaming<
   BedrockAdditionalChatOptions,
   ToolCallLLMMessageOptions
@@ -151,174 +133,6 @@ export const TOOL_CALL_MODELS = [
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
 ];
 
-abstract class Provider<ProviderStreamEvent extends {} = {}> {
-  abstract getTextFromResponse(response: Record<string, any>): string;
-
-  abstract getToolsFromResponse<T extends {} = {}>(
-    response: Record<string, any>,
-  ): T[];
-
-  getStreamingEventResponse(
-    response: Record<string, any>,
-  ): ProviderStreamEvent | undefined {
-    return response.chunk?.bytes
-      ? (JSON.parse(toUtf8(response.chunk?.bytes)) as ProviderStreamEvent)
-      : undefined;
-  }
-
-  async *reduceStream(
-    stream: AsyncIterable<ResponseStream>,
-  ): BedrockChatStreamResponse {
-    yield* streamConverter(stream, (response) => {
-      return {
-        delta: this.getTextFromStreamResponse(response),
-        raw: response,
-      };
-    });
-  }
-
-  getTextFromStreamResponse(response: Record<string, any>): string {
-    return this.getTextFromResponse(response);
-  }
-
-  abstract getRequestBody<T extends ChatMessage>(
-    metadata: LLMMetadata,
-    messages: T[],
-    tools?: BaseTool[],
-    options?: BedrockAdditionalChatOptions,
-  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput;
-}
-
-class AnthropicProvider extends Provider<StreamEvent> {
-  getResultFromResponse(
-    response: Record<string, any>,
-  ): AnthropicNoneStreamingResponse {
-    return JSON.parse(toUtf8(response.body));
-  }
-
-  getToolsFromResponse<AnthropicToolContent>(
-    response: Record<string, any>,
-  ): AnthropicToolContent[] {
-    const result = this.getResultFromResponse(response);
-    return result.content
-      .filter((item) => item.type === "tool_use")
-      .map((item) => item as AnthropicToolContent);
-  }
-
-  getTextFromResponse(response: Record<string, any>): string {
-    const result = this.getResultFromResponse(response);
-    return result.content
-      .filter((item) => item.type === "text")
-      .map((item) => (item as AnthropicTextContent).text)
-      .join(" ");
-  }
-
-  getTextFromStreamResponse(response: Record<string, any>): string {
-    const event = this.getStreamingEventResponse(response);
-    if (event?.type === "content_block_delta") {
-      if (event.delta.type === "text_delta") return event.delta.text;
-      if (event.delta.type === "input_json_delta")
-        return event.delta.partial_json;
-    }
-    return "";
-  }
-
-  async *reduceStream(
-    stream: AsyncIterable<ResponseStream>,
-  ): BedrockChatStreamResponse {
-    let collecting = [];
-    let tool: ToolBlock | undefined = undefined;
-    // #TODO this should be broken down into a separate consumer
-    for await (const response of stream) {
-      const event = this.getStreamingEventResponse(response);
-      if (
-        event?.type === "content_block_start" &&
-        event.content_block.type === "tool_use"
-      ) {
-        tool = event.content_block;
-        continue;
-      }
-
-      if (
-        event?.type === "content_block_delta" &&
-        event.delta.type === "input_json_delta"
-      ) {
-        collecting.push(event.delta.partial_json);
-      }
-
-      let options: undefined | ToolCallLLMMessageOptions = undefined;
-      if (tool && collecting.length) {
-        const input = collecting.filter((item) => item).join("");
-        // We have all we need to parse the tool_use json
-        if (event?.type === "content_block_stop") {
-          options = {
-            toolCall: [
-              {
-                id: tool.id,
-                name: tool.name,
-                input: JSON.parse(input),
-              } as ToolCall,
-            ],
-          };
-          // reset the collection/tool
-          collecting = [];
-          tool = undefined;
-        } else {
-          options = {
-            toolCall: [
-              {
-                id: tool.id,
-                name: tool.name,
-                input,
-              } as PartialToolCall,
-            ],
-          };
-        }
-      }
-      const delta = this.getTextFromStreamResponse(response);
-      if (!delta && !options) continue;
-
-      yield {
-        delta,
-        options,
-        raw: response,
-      };
-    }
-  }
-
-  getRequestBody<T extends ChatMessage<ToolCallLLMMessageOptions>>(
-    metadata: LLMMetadata,
-    messages: T[],
-    tools?: BaseTool[],
-    options?: BedrockAdditionalChatOptions,
-  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
-    const extra: Record<string, unknown> = {};
-    if (options?.toolChoice) {
-      extra["tool_choice"] = options?.toolChoice;
-    }
-    const mapped = mapChatMessagesToAnthropicMessages(messages);
-    return {
-      modelId: metadata.model,
-      contentType: "application/json",
-      accept: "application/json",
-      body: JSON.stringify({
-        anthropic_version: "bedrock-2023-05-31",
-        messages: mapped,
-        tools: mapBaseToolsToAnthropicTools(tools),
-        max_tokens: metadata.maxTokens,
-        temperature: metadata.temperature,
-        top_p: metadata.topP,
-        ...extra,
-      }),
-    };
-  }
-}
-
-// Other providers could go here
-const PROVIDERS: { [key: string]: Provider } = {
-  anthropic: new AnthropicProvider(),
-};
-
 const getProvider = (model: string): Provider => {
   const providerName = model.split(".")[0];
   if (!(providerName in PROVIDERS)) {
@@ -336,6 +150,18 @@ export type BedrockModelParams = {
   maxTokens?: number;
 };
 
+export const BEDROCK_MODEL_MAX_TOKENS: Partial<Record<BEDROCK_MODELS, number>> =
+  {
+    [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_SONNET]: 4096,
+    [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_HAIKU]: 4096,
+    [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_OPUS]: 4096,
+    [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET]: 4096,
+    [BEDROCK_MODELS.META_LLAMA2_13B_CHAT]: 2048,
+    [BEDROCK_MODELS.META_LLAMA2_70B_CHAT]: 2048,
+    [BEDROCK_MODELS.META_LLAMA3_8B_INSTRUCT]: 2048,
+    [BEDROCK_MODELS.META_LLAMA3_70B_INSTRUCT]: 2048,
+  };
+
 const DEFAULT_BEDROCK_PARAMS = {
   temperature: 0.1,
   topP: 1,
@@ -373,6 +199,10 @@ export class Bedrock extends ToolCallLLM<BedrockAdditionalChatOptions> {
     this.temperature = temperature ?? DEFAULT_BEDROCK_PARAMS.temperature;
     this.topP = topP ?? DEFAULT_BEDROCK_PARAMS.topP;
     this.client = new BedrockRuntimeClient(params);
+
+    if (!this.supportToolCall) {
+      console.warn(`The model "${this.model}" doesn't support ToolCall`);
+    }
   }
 
   get supportToolCall(): boolean {
@@ -402,10 +232,13 @@ export class Bedrock extends ToolCallLLM<BedrockAdditionalChatOptions> {
     );
     const command = new InvokeModelCommand(input);
     const response = await this.client.send(command);
-    const tools = this.provider.getToolsFromResponse(response);
-    const options: ToolCallLLMMessageOptions = tools.length
-      ? { toolCall: tools }
-      : {};
+    let options: ToolCallLLMMessageOptions = {};
+    if (this.supportToolCall) {
+      const tools = this.provider.getToolsFromResponse(response);
+      if (tools.length) {
+        options = { toolCall: tools };
+      }
+    }
     return {
       raw: response,
       message: {

packages/community/src/llm/bedrock/provider.ts

@@ -0,0 +1,59 @@
+import {
+  type InvokeModelCommandInput,
+  type InvokeModelWithResponseStreamCommandInput,
+  ResponseStream,
+} from "@aws-sdk/client-bedrock-runtime";
+import {
+  type BaseTool,
+  type ChatMessage,
+  type ChatResponseChunk,
+  type LLMMetadata,
+  type ToolCallLLMMessageOptions,
+} from "@llamaindex/core/llms";
+import { streamConverter } from "@llamaindex/core/utils";
+import type { ToolChoice } from "./types";
+import { toUtf8 } from "./utils";
+
+export type BedrockAdditionalChatOptions = { toolChoice: ToolChoice };
+
+export type BedrockChatStreamResponse = AsyncIterable<
+  ChatResponseChunk<ToolCallLLMMessageOptions>
+>;
+
+export abstract class Provider<ProviderStreamEvent extends {} = {}> {
+  abstract getTextFromResponse(response: Record<string, any>): string;
+
+  abstract getToolsFromResponse<T extends {} = {}>(
+    response: Record<string, any>,
+  ): T[];
+
+  getStreamingEventResponse(
+    response: Record<string, any>,
+  ): ProviderStreamEvent | undefined {
+    return response.chunk?.bytes
+      ? (JSON.parse(toUtf8(response.chunk?.bytes)) as ProviderStreamEvent)
+      : undefined;
+  }
+
+  async *reduceStream(
+    stream: AsyncIterable<ResponseStream>,
+  ): BedrockChatStreamResponse {
+    yield* streamConverter(stream, (response) => {
+      return {
+        delta: this.getTextFromStreamResponse(response),
+        raw: response,
+      };
+    });
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    return this.getTextFromResponse(response);
+  }
+
+  abstract getRequestBody<T extends ChatMessage>(
+    metadata: LLMMetadata,
+    messages: T[],
+    tools?: BaseTool[],
+    options?: BedrockAdditionalChatOptions,
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput;
+}

packages/community/src/llm/bedrock/providers/anthropic.ts

@@ -0,0 +1,154 @@
+import {
+  type InvokeModelCommandInput,
+  type InvokeModelWithResponseStreamCommandInput,
+  ResponseStream,
+} from "@aws-sdk/client-bedrock-runtime";
+import type {
+  BaseTool,
+  ChatMessage,
+  LLMMetadata,
+  PartialToolCall,
+  ToolCall,
+  ToolCallLLMMessageOptions,
+} from "@llamaindex/core/llms";
+import {
+  type BedrockAdditionalChatOptions,
+  type BedrockChatStreamResponse,
+  Provider,
+} from "../provider";
+import type {
+  AnthropicNoneStreamingResponse,
+  AnthropicStreamEvent,
+  AnthropicTextContent,
+  ToolBlock,
+} from "../types";
+import {
+  mapBaseToolsToAnthropicTools,
+  mapChatMessagesToAnthropicMessages,
+  toUtf8,
+} from "../utils";
+
+export class AnthropicProvider extends Provider<AnthropicStreamEvent> {
+  getResultFromResponse(
+    response: Record<string, any>,
+  ): AnthropicNoneStreamingResponse {
+    return JSON.parse(toUtf8(response.body));
+  }
+
+  getToolsFromResponse<AnthropicToolContent>(
+    response: Record<string, any>,
+  ): AnthropicToolContent[] {
+    const result = this.getResultFromResponse(response);
+    return result.content
+      .filter((item) => item.type === "tool_use")
+      .map((item) => item as AnthropicToolContent);
+  }
+
+  getTextFromResponse(response: Record<string, any>): string {
+    const result = this.getResultFromResponse(response);
+    return result.content
+      .filter((item) => item.type === "text")
+      .map((item) => (item as AnthropicTextContent).text)
+      .join(" ");
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    const event = this.getStreamingEventResponse(response);
+    if (event?.type === "content_block_delta") {
+      if (event.delta.type === "text_delta") return event.delta.text;
+      if (event.delta.type === "input_json_delta")
+        return event.delta.partial_json;
+    }
+    return "";
+  }
+
+  async *reduceStream(
+    stream: AsyncIterable<ResponseStream>,
+  ): BedrockChatStreamResponse {
+    let collecting = [];
+    let tool: ToolBlock | undefined = undefined;
+    // #TODO this should be broken down into a separate consumer
+    for await (const response of stream) {
+      const event = this.getStreamingEventResponse(response);
+      if (
+        event?.type === "content_block_start" &&
+        event.content_block.type === "tool_use"
+      ) {
+        tool = event.content_block;
+        continue;
+      }
+
+      if (
+        event?.type === "content_block_delta" &&
+        event.delta.type === "input_json_delta"
+      ) {
+        collecting.push(event.delta.partial_json);
+      }
+
+      let options: undefined | ToolCallLLMMessageOptions = undefined;
+      if (tool && collecting.length) {
+        const input = collecting.filter((item) => item).join("");
+        // We have all we need to parse the tool_use json
+        if (event?.type === "content_block_stop") {
+          options = {
+            toolCall: [
+              {
+                id: tool.id,
+                name: tool.name,
+                input: JSON.parse(input),
+              } as ToolCall,
+            ],
+          };
+          // reset the collection/tool
+          collecting = [];
+          tool = undefined;
+        } else {
+          options = {
+            toolCall: [
+              {
+                id: tool.id,
+                name: tool.name,
+                input,
+              } as PartialToolCall,
+            ],
+          };
+        }
+      }
+      const delta = this.getTextFromStreamResponse(response);
+      if (!delta && !options) continue;
+
+      yield {
+        delta,
+        options,
+        raw: response,
+      };
+    }
+  }
+
+  getRequestBody<T extends ChatMessage<ToolCallLLMMessageOptions>>(
+    metadata: LLMMetadata,
+    messages: T[],
+    tools?: BaseTool[],
+    options?: BedrockAdditionalChatOptions,
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
+    const extra: Record<string, unknown> = {};
+    if (options?.toolChoice) {
+      extra["tool_choice"] = options?.toolChoice;
+    }
+    const mapped = mapChatMessagesToAnthropicMessages(messages);
+    return {
+      modelId: metadata.model,
+      contentType: "application/json",
+      accept: "application/json",
+      body: JSON.stringify({
+        anthropic_version: "bedrock-2023-05-31",
+        messages: mapped,
+        tools: mapBaseToolsToAnthropicTools(tools),
+        max_tokens: metadata.maxTokens,
+        temperature: metadata.temperature,
+        top_p: metadata.topP,
+        ...extra,
+      }),
+    };
+  }
+}

packages/community/src/llm/bedrock/providers/index.ts

@@ -0,0 +1,9 @@
+import { Provider } from "../provider";
+import { AnthropicProvider } from "./anthropic";
+import { MetaProvider } from "./meta";
+
+// Other providers should go here
+export const PROVIDERS: { [key: string]: Provider } = {
+  anthropic: new AnthropicProvider(),
+  meta: new MetaProvider(),
+};

packages/community/src/llm/bedrock/providers/meta.ts

@@ -0,0 +1,69 @@
+import type {
+  InvokeModelCommandInput,
+  InvokeModelWithResponseStreamCommandInput,
+} from "@aws-sdk/client-bedrock-runtime";
+import type { ChatMessage, LLMMetadata } from "@llamaindex/core/llms";
+import type { MetaNoneStreamingResponse, MetaStreamEvent } from "../types";
+import {
+  mapChatMessagesToMetaLlama2Messages,
+  mapChatMessagesToMetaLlama3Messages,
+  toUtf8,
+} from "../utils";
+
+import { Provider } from "../provider";
+
+export class MetaProvider extends Provider<MetaStreamEvent> {
+  constructor() {
+    super();
+  }
+
+  getResultFromResponse(
+    response: Record<string, any>,
+  ): MetaNoneStreamingResponse {
+    return JSON.parse(toUtf8(response.body));
+  }
+
+  getToolsFromResponse(_response: Record<string, any>): never {
+    throw new Error("Not supported by this provider.");
+  }
+
+  getTextFromResponse(response: Record<string, any>): string {
+    const result = this.getResultFromResponse(response);
+    return result.generation;
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    const event = this.getStreamingEventResponse(response);
+    if (event?.generation) {
+      return event.generation;
+    }
+    return "";
+  }
+
+  getRequestBody<T extends ChatMessage>(
+    metadata: LLMMetadata,
+    messages: T[],
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
+    let promptFunction: (messages: ChatMessage[]) => string;
+
+    if (metadata.model.startsWith("meta.llama3")) {
+      promptFunction = mapChatMessagesToMetaLlama3Messages;
+    } else if (metadata.model.startsWith("meta.llama2")) {
+      promptFunction = mapChatMessagesToMetaLlama2Messages;
+    } else {
+      throw new Error(`Meta model ${metadata.model} is not supported`);
+    }
+
+    return {
+      modelId: metadata.model,
+      contentType: "application/json",
+      accept: "application/json",
+      body: JSON.stringify({
+        prompt: promptFunction(messages),
+        max_gen_len: metadata.maxTokens,
+        temperature: metadata.temperature,
+        top_p: metadata.topP,
+      }),
+    };
+  }
+}

packages/community/src/llm/bedrock/types.ts

@@ -79,7 +79,7 @@ export type ToolChoice =
   | { type: "auto" }
   | { type: "tool"; name: string };
 
-export type StreamEvent =
+export type AnthropicStreamEvent =
   | { type: "message_start"; message: Message }
   | ContentBlockStart
   | ContentBlockDelta
@@ -93,6 +93,8 @@ export type AnthropicContent =
   | AnthropicToolContent
   | AnthropicToolResultContent;
 
+export type MetaTextContent = string;
+
 export type AnthropicTextContent = {
   type: "text";
   text: string;
@@ -133,6 +135,11 @@ export type AnthropicMessage = {
   content: AnthropicContent[];
 };
 
+export type MetaMessage = {
+  role: "user" | "assistant" | "system";
+  content: MetaTextContent;
+};
+
 export type AnthropicNoneStreamingResponse = {
   id: string;
   type: "message";
@@ -143,3 +150,16 @@ export type AnthropicNoneStreamingResponse = {
   stop_sequence?: string;
   usage: { input_tokens: number; output_tokens: number };
 };
+
+type MetaResponse = {
+  generation: string;
+  prompt_token_count: number;
+  generation_token_count: number;
+  stop_reason: "stop" | "length";
+};
+
+export type MetaStreamEvent = MetaResponse & {
+  "amazon-bedrock-invocationMetrics": InvocationMetrics;
+};
+
+export type MetaNoneStreamingResponse = MetaResponse;

packages/community/src/llm/bedrock/utils.ts

@@ -1,18 +1,20 @@
+import type { JSONObject } from "@llamaindex/core/global";
 import type {
   BaseTool,
   ChatMessage,
-  JSONObject,
   MessageContent,
   MessageContentDetail,
+  MessageContentTextDetail,
   ToolCallLLMMessageOptions,
   ToolMetadata,
-} from "llamaindex";
+} from "@llamaindex/core/llms";
 import type {
   AnthropicContent,
   AnthropicImageContent,
   AnthropicMediaTypes,
   AnthropicMessage,
   AnthropicTextContent,
+  MetaMessage,
 } from "./types.js";
 
 const ACCEPTED_IMAGE_MIME_TYPES = [
@@ -148,6 +150,85 @@ export const mapChatMessagesToAnthropicMessages = <
   return mergeNeighboringSameRoleMessages(mapped);
 };
 
+export const mapChatMessagesToMetaMessages = <T extends ChatMessage>(
+  messages: T[],
+): MetaMessage[] => {
+  return messages.map((msg) => {
+    let content: string = "";
+    if (typeof msg.content === "string") {
+      content = msg.content;
+    } else if (msg.content.length) {
+      content = (msg.content[0] as MessageContentTextDetail).text;
+    }
+    return {
+      role:
+        msg.role === "assistant"
+          ? "assistant"
+          : msg.role === "user"
+            ? "user"
+            : "system",
+      content,
+    };
+  });
+};
+
+/**
+ * Documentation at https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-3
+ */
+export const mapChatMessagesToMetaLlama3Messages = <T extends ChatMessage>(
+  messages: T[],
+): string => {
+  const mapped = mapChatMessagesToMetaMessages(messages).map((message) => {
+    const text = message.content;
+    return `<|start_header_id|>${message.role}<|end_header_id|>\n${text}\n<|eot_id|>\n`;
+  });
+  return (
+    "<|begin_of_text|>" +
+    mapped.join("\n") +
+    "\n<|start_header_id|>assistant<|end_header_id|>\n"
+  );
+};
+
+/**
+ * Documentation at https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-2
+ */
+export const mapChatMessagesToMetaLlama2Messages = <T extends ChatMessage>(
+  messages: T[],
+): string => {
+  const mapped = mapChatMessagesToMetaMessages(messages);
+  let output = "<s>";
+  let insideInst = false;
+  let needsStartAgain = false;
+  for (const message of mapped) {
+    if (needsStartAgain) {
+      output += "<s>";
+      needsStartAgain = false;
+    }
+    const text = message.content;
+    if (message.role === "system") {
+      if (!insideInst) {
+        output += "[INST] ";
+        insideInst = true;
+      }
+      output += `<<SYS>>\n${text}\n<</SYS>>\n`;
+    } else if (message.role === "user") {
+      output += text;
+      if (insideInst) {
+        output += " [/INST]";
+        insideInst = false;
+      }
+    } else if (message.role === "assistant") {
+      if (insideInst) {
+        output += " [/INST]";
+        insideInst = false;
+      }
+      output += ` ${text} </s>\n`;
+      needsStartAgain = true;
+    }
+  }
+  return output;
+};
+
 export const mapTextContent = (text: string): AnthropicTextContent => {
   return { type: "text", text };
 };

packages/community/tsconfig.script.json

@@ -1,9 +0,0 @@
-{
-  "extends": "../../tsconfig.json",
-  "compilerOptions": {
-    "outDir": "./dist/script/type",
-    "tsBuildInfoFile": "./dist/script/.tsbuildinfo",
-    "emitDeclarationOnly": true
-  },
-  "include": ["./tsup.config.ts"]
-}

packages/community/tsup.config.ts

@@ -1,9 +0,0 @@
-import { defineConfig } from "tsup";
-
-export default defineConfig([
-  {
-    entry: ["src/index.ts", "src/llm/bedrock/base.ts"],
-    format: ["cjs", "esm"],
-    sourcemap: true,
-  },
-]);

packages/core/CHANGELOG.md

@@ -1,5 +1,28 @@
 # @llamaindex/core
 
+## 0.1.0
+
+### Minor Changes
+
+- 16ef5dd: refactor: simplify callback manager
+
+  Change `event.detail.payload` to `event.detail`
+
+### Patch Changes
+
+- 16ef5dd: refactor: move callback manager & llm to core module
+
+  For people who import `llamaindex/llms/base` or `llamaindex/llms/utils`,
+  use `@llamaindex/core/llms` and `@llamaindex/core/utils` instead.
+
+## 0.0.3
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - @llamaindex/env@0.1.8
+
 ## 0.0.2
 
 ### Patch Changes

packages/core/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/core",
   "type": "module",
-  "version": "0.0.2",
+  "version": "0.1.0",
   "description": "LlamaIndex Core Module",
   "exports": {
     "./llms": {
@@ -59,6 +59,20 @@
         "types": "./dist/schema/index.d.ts",
         "default": "./dist/schema/index.js"
       }
+    },
+    "./utils": {
+      "require": {
+        "types": "./dist/utils/index.d.cts",
+        "default": "./dist/utils/index.cjs"
+      },
+      "import": {
+        "types": "./dist/utils/index.d.ts",
+        "default": "./dist/utils/index.js"
+      },
+      "default": {
+        "types": "./dist/utils/index.d.ts",
+        "default": "./dist/utils/index.js"
+      }
     }
   },
   "files": [
@@ -75,7 +89,7 @@
   },
   "devDependencies": {
     "ajv": "^8.16.0",
-    "bunchee": "^5.2.1"
+    "bunchee": "5.3.0-beta.0"
   },
   "dependencies": {
     "@llamaindex/env": "workspace:*",

packages/core/src/global/index.ts

@@ -1 +1,11 @@
 export { Settings } from "./settings";
+export { CallbackManager } from "./settings/callback-manager";
+export type {
+  LLMEndEvent,
+  LLMStartEvent,
+  LLMStreamEvent,
+  LLMToolCallEvent,
+  LLMToolResultEvent,
+  LlamaIndexEventMaps,
+} from "./settings/callback-manager";
+export type { JSONArray, JSONObject, JSONValue } from "./type";

packages/core/src/global/settings.ts

@@ -1,3 +1,9 @@
+import {
+  type CallbackManager,
+  getCallbackManager,
+  setCallbackManager,
+  withCallbackManager,
+} from "./settings/callback-manager";
 import {
   getChunkSize,
   setChunkSize,
@@ -14,4 +20,19 @@ export const Settings = {
   withChunkSize<Result>(chunkSize: number, fn: () => Result): Result {
     return withChunkSize(chunkSize, fn);
   },
+
+  get callbackManager(): CallbackManager {
+    return getCallbackManager();
+  },
+
+  set callbackManager(callbackManager: CallbackManager) {
+    setCallbackManager(callbackManager);
+  },
+
+  withCallbackManager<Result>(
+    callbackManager: CallbackManager,
+    fn: () => Result,
+  ): Result {
+    return withCallbackManager(callbackManager, fn);
+  },
 };

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

@@ -0,0 +1,140 @@
+import { AsyncLocalStorage, CustomEvent } from "@llamaindex/env";
+import type {
+  ChatMessage,
+  ChatResponse,
+  ChatResponseChunk,
+  ToolCall,
+  ToolOutput,
+} from "../../llms";
+import { EventCaller, getEventCaller } from "../../utils/event-caller";
+import type { UUID } from "../type";
+
+export type LLMStartEvent = {
+  id: UUID;
+  messages: ChatMessage[];
+};
+
+export type LLMToolCallEvent = {
+  toolCall: ToolCall;
+};
+
+export type LLMToolResultEvent = {
+  toolCall: ToolCall;
+  toolResult: ToolOutput;
+};
+
+export type LLMEndEvent = {
+  id: UUID;
+  response: ChatResponse;
+};
+
+export type LLMStreamEvent = {
+  id: UUID;
+  chunk: ChatResponseChunk;
+};
+
+export interface LlamaIndexEventMaps {
+  "llm-start": LLMStartEvent;
+  "llm-end": LLMEndEvent;
+  "llm-tool-call": LLMToolCallEvent;
+  "llm-tool-result": LLMToolResultEvent;
+  "llm-stream": LLMStreamEvent;
+}
+
+export class LlamaIndexCustomEvent<T = any> extends CustomEvent<T> {
+  reason: EventCaller | null = null;
+  private constructor(
+    event: string,
+    options?: CustomEventInit & {
+      reason?: EventCaller | null;
+    },
+  ) {
+    super(event, options);
+    this.reason = options?.reason ?? null;
+  }
+
+  static fromEvent<Type extends keyof LlamaIndexEventMaps>(
+    type: Type,
+    detail: LlamaIndexEventMaps[Type],
+  ) {
+    return new LlamaIndexCustomEvent(type, {
+      detail: detail,
+      reason: getEventCaller(),
+    });
+  }
+}
+
+type EventHandler<Event> = (event: LlamaIndexCustomEvent<Event>) => void;
+
+export class CallbackManager {
+  #handlers = new Map<keyof LlamaIndexEventMaps, EventHandler<any>[]>();
+
+  on<K extends keyof LlamaIndexEventMaps>(
+    event: K,
+    handler: EventHandler<LlamaIndexEventMaps[K]>,
+  ) {
+    if (!this.#handlers.has(event)) {
+      this.#handlers.set(event, []);
+    }
+    this.#handlers.get(event)!.push(handler);
+    return this;
+  }
+
+  off<K extends keyof LlamaIndexEventMaps>(
+    event: K,
+    handler: EventHandler<LlamaIndexEventMaps[K]>,
+  ) {
+    if (!this.#handlers.has(event)) {
+      return this;
+    }
+    const cbs = this.#handlers.get(event)!;
+    const index = cbs.indexOf(handler);
+    if (index > -1) {
+      cbs.splice(index, 1);
+    }
+    return this;
+  }
+
+  dispatchEvent<K extends keyof LlamaIndexEventMaps>(
+    event: K,
+    detail: LlamaIndexEventMaps[K],
+  ) {
+    const cbs = this.#handlers.get(event);
+    if (!cbs) {
+      return;
+    }
+    queueMicrotask(() => {
+      cbs.forEach((handler) =>
+        handler(
+          LlamaIndexCustomEvent.fromEvent(event, structuredClone(detail)),
+        ),
+      );
+    });
+  }
+}
+
+export const globalCallbackManager = new CallbackManager();
+
+const callbackManagerAsyncLocalStorage =
+  new AsyncLocalStorage<CallbackManager>();
+
+let currentCallbackManager: CallbackManager | null = null;
+
+export function getCallbackManager(): CallbackManager {
+  return (
+    callbackManagerAsyncLocalStorage.getStore() ??
+    currentCallbackManager ??
+    globalCallbackManager
+  );
+}
+
+export function setCallbackManager(callbackManager: CallbackManager) {
+  currentCallbackManager = callbackManager;
+}
+
+export function withCallbackManager<Result>(
+  callbackManager: CallbackManager,
+  fn: () => Result,
+): Result {
+  return callbackManagerAsyncLocalStorage.run(callbackManager, fn);
+}

packages/core/src/llms/base.ts

@@ -1,3 +1,5 @@
+import { streamConverter } from "../utils";
+import { extractText } from "../utils/llms";
 import type {
   ChatResponse,
   ChatResponseChunk,
@@ -9,8 +11,7 @@ import type {
   LLMCompletionParamsStreaming,
   LLMMetadata,
   ToolCallLLMMessageOptions,
-} from "@llamaindex/core/llms";
-import { extractText, streamConverter } from "./utils.js";
+} from "./type";
 
 export abstract class BaseLLM<
   AdditionalChatOptions extends object = object,

packages/core/src/llms/index.ts

@@ -1,3 +1,4 @@
+export { BaseLLM, ToolCallLLM } from "./base";
 export type {
   BaseTool,
   BaseToolWithCall,

packages/core/src/utils/event-caller.ts

@@ -1,5 +1,14 @@
 import { AsyncLocalStorage, randomUUID } from "@llamaindex/env";
-import { isAsyncIterable, isIterable } from "../utils.js";
+
+export const isAsyncIterable = (
+  obj: unknown,
+): obj is AsyncIterable<unknown> => {
+  return obj != null && typeof obj === "object" && Symbol.asyncIterator in obj;
+};
+
+export const isIterable = (obj: unknown): obj is Iterable<unknown> => {
+  return obj != null && typeof obj === "object" && Symbol.iterator in obj;
+};
 
 const eventReasonAsyncLocalStorage = new AsyncLocalStorage<EventCaller>();
 

packages/core/src/utils/index.ts

@@ -0,0 +1,54 @@
+export { wrapEventCaller } from "./event-caller";
+
+export async function* streamConverter<S, D>(
+  stream: AsyncIterable<S>,
+  converter: (s: S) => D | null,
+): AsyncIterable<D> {
+  for await (const data of stream) {
+    const newData = converter(data);
+    if (newData === null) {
+      return;
+    }
+    yield newData;
+  }
+}
+
+export async function* streamCallbacks<S>(
+  stream: AsyncIterable<S>,
+  callbacks: {
+    finished?: (value?: S) => void;
+  },
+): AsyncIterable<S> {
+  let value: S | undefined;
+  for await (value of stream) {
+    yield value;
+  }
+  if (callbacks.finished) {
+    callbacks.finished(value);
+  }
+}
+
+export async function* streamReducer<S, D>(params: {
+  stream: AsyncIterable<S>;
+  reducer: (previousValue: D, currentValue: S) => D;
+  initialValue: D;
+  finished?: (value: D) => void;
+}): AsyncIterable<S> {
+  let value = params.initialValue;
+  for await (const data of params.stream) {
+    value = params.reducer(value, data);
+    yield data;
+  }
+  if (params.finished) {
+    params.finished(value);
+  }
+}
+
+export { wrapLLMEvent } from "./wrap-llm-event";
+
+export {
+  extractDataUrlComponents,
+  extractImage,
+  extractSingleText,
+  extractText,
+} from "./llms";

packages/core/src/utils/llms.ts

@@ -0,0 +1,79 @@
+import type {
+  MessageContent,
+  MessageContentDetail,
+  MessageContentTextDetail,
+} from "../llms";
+import type { ImageType } from "../schema";
+
+/**
+ * Extracts just the text from a multi-modal message or the message itself if it's just text.
+ *
+ * @param message The message to extract text from.
+ * @returns The extracted text
+ */
+export function extractText(message: MessageContent): string {
+  if (typeof message !== "string" && !Array.isArray(message)) {
+    console.warn(
+      "extractText called with non-MessageContent message, this is likely a bug.",
+    );
+    return `${message}`;
+  } else if (typeof message !== "string" && Array.isArray(message)) {
+    // message is of type MessageContentDetail[] - retrieve just the text parts and concatenate them
+    // so we can pass them to the context generator
+    return message
+      .filter((c): c is MessageContentTextDetail => c.type === "text")
+      .map((c) => c.text)
+      .join("\n\n");
+  } else {
+    return message;
+  }
+}
+
+/**
+ * Extracts a single text from a multi-modal message content
+ *
+ * @param message The message to extract images from.
+ * @returns The extracted images
+ */
+export function extractSingleText(
+  message: MessageContentDetail,
+): string | null {
+  if (message.type === "text") {
+    return message.text;
+  }
+  return null;
+}
+
+/**
+ * Extracts an image from a multi-modal message content
+ *
+ * @param message The message to extract images from.
+ * @returns The extracted images
+ */
+export function extractImage(message: MessageContentDetail): ImageType | null {
+  if (message.type === "image_url") {
+    return new URL(message.image_url.url);
+  }
+  return null;
+}
+
+export const extractDataUrlComponents = (
+  dataUrl: string,
+): {
+  mimeType: string;
+  base64: string;
+} => {
+  const parts = dataUrl.split(";base64,");
+
+  if (parts.length !== 2 || !parts[0].startsWith("data:")) {
+    throw new Error("Invalid data URL");
+  }
+
+  const mimeType = parts[0].slice(5);
+  const base64 = parts[1];
+
+  return {
+    mimeType,
+    base64,
+  };
+};

packages/core/src/utils/wrap-llm-event.ts

@@ -0,0 +1,80 @@
+import { AsyncLocalStorage, randomUUID } from "@llamaindex/env";
+import { getCallbackManager } from "../global/settings/callback-manager";
+import type { ChatResponse, ChatResponseChunk, LLM, LLMChat } from "../llms";
+
+export function wrapLLMEvent<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+>(
+  originalMethod: LLMChat<
+    AdditionalChatOptions,
+    AdditionalMessageOptions
+  >["chat"],
+  _context: ClassMethodDecoratorContext,
+) {
+  return async function withLLMEvent(
+    this: LLM<AdditionalChatOptions, AdditionalMessageOptions>,
+    ...params: Parameters<
+      LLMChat<AdditionalChatOptions, AdditionalMessageOptions>["chat"]
+    >
+  ): ReturnType<
+    LLMChat<AdditionalChatOptions, AdditionalMessageOptions>["chat"]
+  > {
+    const id = randomUUID();
+    getCallbackManager().dispatchEvent("llm-start", {
+      id,
+      messages: params[0].messages,
+    });
+    const response = await originalMethod.call(this, ...params);
+    if (Symbol.asyncIterator in response) {
+      // save snapshot to restore it after the response is done
+      const snapshot = AsyncLocalStorage.snapshot();
+      const originalAsyncIterator = {
+        [Symbol.asyncIterator]: response[Symbol.asyncIterator].bind(response),
+      };
+      response[Symbol.asyncIterator] = async function* () {
+        const finalResponse = {
+          raw: [] as ChatResponseChunk[],
+          message: {
+            content: "",
+            role: "assistant",
+            options: {},
+          },
+        } satisfies ChatResponse;
+        let firstOne = false;
+        for await (const chunk of originalAsyncIterator) {
+          if (!firstOne) {
+            firstOne = true;
+            finalResponse.message.content = chunk.delta;
+          } else {
+            finalResponse.message.content += chunk.delta;
+          }
+          if (chunk.options) {
+            finalResponse.message.options = {
+              ...finalResponse.message.options,
+              ...chunk.options,
+            };
+          }
+          getCallbackManager().dispatchEvent("llm-stream", {
+            id,
+            chunk,
+          });
+          finalResponse.raw.push(chunk);
+          yield chunk;
+        }
+        snapshot(() => {
+          getCallbackManager().dispatchEvent("llm-end", {
+            id,
+            response: finalResponse,
+          });
+        });
+      };
+    } else {
+      getCallbackManager().dispatchEvent("llm-end", {
+        id,
+        response,
+      });
+    }
+    return response;
+  };
+}

packages/core/tests/event-system.test.ts

@@ -0,0 +1,67 @@
+import { CallbackManager, Settings } from "@llamaindex/core/global";
+import { beforeEach, describe, expect, expectTypeOf, test, vi } from "vitest";
+
+declare module "@llamaindex/core/global" {
+  interface LlamaIndexEventMaps {
+    test: {
+      value: number;
+    };
+  }
+}
+
+describe("event system", () => {
+  beforeEach(() => {
+    Settings.callbackManager = new CallbackManager();
+  });
+
+  test("type system", () => {
+    Settings.callbackManager.on("test", (event) => {
+      const data = event.detail;
+      expectTypeOf(data).not.toBeAny();
+      expectTypeOf(data).toEqualTypeOf<{
+        value: number;
+      }>();
+    });
+  });
+
+  test("dispatch event", async () => {
+    let callback;
+    Settings.callbackManager.on(
+      "test",
+      (callback = vi.fn((event) => {
+        const data = event.detail;
+        expect(data.value).toBe(42);
+      })),
+    );
+
+    Settings.callbackManager.dispatchEvent("test", {
+      value: 42,
+    });
+    expect(callback).toHaveBeenCalledTimes(0);
+    await new Promise((resolve) => process.nextTick(resolve));
+    expect(callback).toHaveBeenCalledTimes(1);
+  });
+
+  // rollup doesn't support decorators for now
+  // test('wrap event caller', async () => {
+  //   class A {
+  //     @wrapEventCaller
+  //     fn() {
+  //       Settings.callbackManager.dispatchEvent('test', {
+  //         value: 42
+  //       });
+  //     }
+  //   }
+  //   const a = new A();
+  //   let callback;
+  //   Settings.callbackManager.on('test', callback = vi.fn((event) => {
+  //     const data = event.detail;
+  //     expect(event.reason!.caller).toBe(a);
+  //     expect(data.value).toBe(42);
+  //   }));
+  //   a.fn();
+  //   expect(callback).toHaveBeenCalledTimes(0);
+  //   await new Promise((resolve) => process.nextTick(resolve));
+  //   expect(callback).toHaveBeenCalledTimes(1);
+  // })
+});

packages/core/tsconfig.json

@@ -8,6 +8,7 @@
     "moduleResolution": "Bundler",
     "skipLibCheck": true,
     "strict": true,
+    "lib": ["ESNext", "DOM"],
     "types": ["node"]
   },
   "include": ["./src"],

packages/env/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @llamaindex/env
 
+## 0.1.8
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+
 ## 0.1.7
 
 ### Patch Changes

packages/env/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/env",
   "description": "environment wrapper, supports all JS environment including node, deno, bun, edge runtime, and cloudflare worker",
-  "version": "0.1.7",
+  "version": "0.1.8",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/experimental/CHANGELOG.md

@@ -1,5 +1,49 @@
 # @llamaindex/experimental
 
+## 0.0.52
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.0.51
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.0.50
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.49
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.48
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.47
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.0.46
 
 ### Patch Changes

packages/experimental/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/experimental",
   "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.46",
+  "version": "0.0.52",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/llamaindex/CHANGELOG.md

@@ -1,5 +1,65 @@
 # llamaindex
 
+## 0.5.2
+
+### Patch Changes
+
+- Updated dependencies [3ed6acc]
+  - @llamaindex/cloud@0.2.0
+
+## 0.5.1
+
+### Patch Changes
+
+- 2774681: Add mixedbread's embeddings and reranking API
+- a0f424e: corrected the regex in the react.ts file in extractToolUse & extractJsonStr functions, as mentioned in https://github.com/run-llama/LlamaIndexTS/issues/1019
+
+## 0.5.0
+
+### Minor Changes
+
+- 16ef5dd: refactor: simplify callback manager
+
+  Change `event.detail.payload` to `event.detail`
+
+### Patch Changes
+
+- 16ef5dd: refactor: move callback manager & llm to core module
+
+  For people who import `llamaindex/llms/base` or `llamaindex/llms/utils`,
+  use `@llamaindex/core/llms` and `@llamaindex/core/utils` instead.
+
+- 36ddec4: fix: typo in custom page separator parameter for LlamaParse
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - @llamaindex/core@0.1.0
+  - @llamaindex/cloud@0.1.4
+
+## 0.4.14
+
+### Patch Changes
+
+- Updated dependencies [1c444d5]
+  - @llamaindex/cloud@0.1.3
+
+## 0.4.13
+
+### Patch Changes
+
+- e8f8bea: feat: add boundingBox and targetPages to LlamaParseReader
+- 304484b: feat: add ignoreErrors flag to LlamaParseReader
+
+## 0.4.12
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - @llamaindex/cloud@0.1.2
+  - @llamaindex/core@0.0.3
+  - @llamaindex/env@0.1.8
+
 ## 0.4.11
 
 ### Patch Changes
@@ -101,7 +161,7 @@
 
 ### Patch Changes
 
-- 6bc5bdd: feat: add cache disabling, fast mode, do not unroll columns mode and custom page seperator to LlamaParseReader
+- 6bc5bdd: feat: add cache disabling, fast mode, do not unroll columns mode and custom page separator to LlamaParseReader
 - bf25ff6: fix: polyfill for cloudflare worker
 - e6d6576: chore: use `unpdf`
 

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

@@ -1,5 +1,49 @@
 # @llamaindex/cloudflare-worker-agent-test
 
+## 0.0.36
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.0.35
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.0.34
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.33
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.0.30
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.30",
+  "version": "0.0.36",
   "type": "module",
   "private": true,
   "scripts": {

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

@@ -1,5 +1,49 @@
 # @llamaindex/next-agent-test
 
+## 0.1.36
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.1.35
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.1.34
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.1.33
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.1.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.1.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.1.30
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-agent-test",
-  "version": "0.1.30",
+  "version": "0.1.36",
   "private": true,
   "scripts": {
     "dev": "next dev",

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

@@ -1,5 +1,49 @@
 # test-edge-runtime
 
+## 0.1.35
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.1.34
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.1.33
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.1.32
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.1.31
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.1.30
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.1.29
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.29",
+  "version": "0.1.35",
   "private": true,
   "scripts": {
     "dev": "next dev",

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

@@ -1,5 +1,49 @@
 # @llamaindex/next-node-runtime
 
+## 0.0.17
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.14
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.13
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.0.11
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.0.11",
+  "version": "0.0.17",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-node-runtime/src/actions/openai.ts

@@ -19,10 +19,10 @@ Settings.embedModel = new HuggingFaceEmbedding({
   quantized: false,
 });
 Settings.callbackManager.on("llm-tool-call", (event) => {
-  console.log(event.detail.payload);
+  console.log(event.detail);
 });
 Settings.callbackManager.on("llm-tool-result", (event) => {
-  console.log(event.detail.payload);
+  console.log(event.detail);
 });
 
 export async function getOpenAIModelRequest(query: string) {

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

@@ -1,5 +1,49 @@
 # @llamaindex/waku-query-engine-test
 
+## 0.0.36
+
+### Patch Changes
+
+- llamaindex@0.5.2
+
+## 0.0.35
+
+### Patch Changes
+
+- Updated dependencies [2774681]
+- Updated dependencies [a0f424e]
+  - llamaindex@0.5.1
+
+## 0.0.34
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.33
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
 ## 0.0.30
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.30",
+  "version": "0.0.36",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/e2e/fixtures/llm/openai.ts

@@ -1,3 +1,4 @@
+import { extractText } from "@llamaindex/core/utils";
 import type {
   ChatResponse,
   ChatResponseChunk,
@@ -8,7 +9,6 @@ import type {
   LLMCompletionParamsNonStreaming,
   LLMCompletionParamsStreaming,
 } from "llamaindex";
-import { extractText } from "llamaindex/llm/utils";
 import { deepStrictEqual, strictEqual } from "node:assert";
 import { llmCompleteMockStorage } from "../../node/utils.js";
 

packages/llamaindex/e2e/node/claude.e2e.ts

@@ -1,7 +1,7 @@
+import { extractText } from "@llamaindex/core/utils";
 import { consola } from "consola";
 import { Anthropic, FunctionTool, Settings, type LLM } from "llamaindex";
 import { AnthropicAgent } from "llamaindex/agent/anthropic";
-import { extractText } from "llamaindex/llm/utils";
 import { ok } from "node:assert";
 import { beforeEach, test } from "node:test";
 import { getWeatherTool, sumNumbersTool } from "./fixtures/tools.js";

packages/llamaindex/e2e/node/openai.e2e.ts

@@ -1,3 +1,4 @@
+import { extractText } from "@llamaindex/core/utils";
 import { consola } from "consola";
 import {
   Document,
@@ -14,7 +15,6 @@ import {
   VectorStoreIndex,
   type LLM,
 } from "llamaindex";
-import { extractText } from "llamaindex/llm/utils";
 import { ok, strictEqual } from "node:assert";
 import { readFile } from "node:fs/promises";
 import { join } from "node:path";

packages/llamaindex/e2e/node/react.e2e.ts

@@ -1,5 +1,5 @@
+import { extractText } from "@llamaindex/core/utils";
 import { OpenAI, ReActAgent, Settings, type LLM } from "llamaindex";
-import { extractText } from "llamaindex/llm/utils";
 import { ok } from "node:assert";
 import { beforeEach, test } from "node:test";
 import { getWeatherTool } from "./fixtures/tools.js";

packages/llamaindex/e2e/node/utils.ts

@@ -4,16 +4,17 @@ import {
   type LLMEndEvent,
   type LLMStartEvent,
   type LLMStreamEvent,
-} from "llamaindex";
+} from "@llamaindex/core/global";
+import { CustomEvent } from "@llamaindex/env";
 import { readFile, writeFile } from "node:fs/promises";
 import { join } from "node:path";
 import { type test } from "node:test";
 import { fileURLToPath } from "node:url";
 
 type MockStorage = {
-  llmEventStart: LLMStartEvent["detail"]["payload"][];
-  llmEventEnd: LLMEndEvent["detail"]["payload"][];
-  llmEventStream: LLMStreamEvent["detail"]["payload"][];
+  llmEventStart: LLMStartEvent[];
+  llmEventEnd: LLMEndEvent[];
+  llmEventStream: LLMStreamEvent[];
 };
 
 export const llmCompleteMockStorage: MockStorage = {
@@ -36,35 +37,35 @@ export async function mockLLMEvent(
     llmEventStream: [],
   };
 
-  function captureLLMStart(event: LLMStartEvent) {
-    idMap.set(event.detail.payload.id, `PRESERVE_${counter++}`);
+  function captureLLMStart(event: CustomEvent<LLMStartEvent>) {
+    idMap.set(event.detail.id, `PRESERVE_${counter++}`);
     newLLMCompleteMockStorage.llmEventStart.push({
-      ...event.detail.payload,
+      ...event.detail,
       // @ts-expect-error id is not UUID, but it is fine for testing
       id: idMap.get(event.detail.payload.id)!,
     });
   }
 
-  function captureLLMEnd(event: LLMEndEvent) {
+  function captureLLMEnd(event: CustomEvent<LLMEndEvent>) {
     newLLMCompleteMockStorage.llmEventEnd.push({
-      ...event.detail.payload,
+      ...event.detail,
       // @ts-expect-error id is not UUID, but it is fine for testing
       id: idMap.get(event.detail.payload.id)!,
       response: {
-        ...event.detail.payload.response,
+        ...event.detail.response,
         // hide raw object since it might too big
         raw: null,
       },
     });
   }
 
-  function captureLLMStream(event: LLMStreamEvent) {
+  function captureLLMStream(event: CustomEvent<LLMStreamEvent>) {
     newLLMCompleteMockStorage.llmEventStream.push({
-      ...event.detail.payload,
+      ...event.detail,
       // @ts-expect-error id is not UUID, but it is fine for testing
       id: idMap.get(event.detail.payload.id)!,
       chunk: {
-        ...event.detail.payload.chunk,
+        ...event.detail.chunk,
         // hide raw object since it might too big
         raw: null,
       },

packages/llamaindex/e2e/package.json

@@ -10,6 +10,7 @@
   },
   "devDependencies": {
     "@faker-js/faker": "^8.4.1",
+    "@llamaindex/core": "workspace:*",
     "@types/node": "^20.12.11",
     "consola": "^3.2.3",
     "llamaindex": "workspace:*",

packages/llamaindex/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llamaindex",
-  "version": "0.4.11",
+  "version": "0.5.2",
   "license": "MIT",
   "type": "module",
   "keywords": [
@@ -45,6 +45,7 @@
     "chromadb": "1.8.1",
     "cohere-ai": "7.9.5",
     "groq-sdk": "^0.5.0",
+    "@mixedbread-ai/sdk": "^2.2.11",
     "js-tiktoken": "^1.0.12",
     "lodash": "^4.17.21",
     "magic-bytes.js": "^1.10.0",

packages/llamaindex/src/ChatHistory.ts

@@ -1,9 +1,9 @@
 import type { ChatMessage, LLM, MessageType } from "@llamaindex/core/llms";
+import { extractText } from "@llamaindex/core/utils";
 import { tokenizers, type Tokenizer } from "@llamaindex/env";
 import type { SummaryPrompt } from "./Prompt.js";
 import { defaultSummaryPrompt, messagesToHistoryStr } from "./Prompt.js";
 import { OpenAI } from "./llm/openai.js";
-import { extractText } from "./llm/utils.js";
 
 /**
  * A ChatHistory is used to keep the state of back and forth chat messages

packages/llamaindex/src/EngineResponse.ts

@@ -4,7 +4,7 @@ import type {
   ChatResponseChunk,
 } from "@llamaindex/core/llms";
 import type { NodeWithScore } from "@llamaindex/core/schema";
-import { extractText } from "./llm/utils.js";
+import { extractText } from "@llamaindex/core/utils";
 
 export class EngineResponse implements ChatResponse, ChatResponseChunk {
   sourceNodes?: NodeWithScore[];

packages/llamaindex/src/Settings.ts

@@ -1,5 +1,7 @@
-import { Settings as CoreSettings } from "@llamaindex/core/global";
-import { CallbackManager } from "./callbacks/CallbackManager.js";
+import {
+  type CallbackManager,
+  Settings as CoreSettings,
+} from "@llamaindex/core/global";
 import { OpenAI } from "./llm/openai.js";
 
 import { PromptHelper } from "./PromptHelper.js";
@@ -9,11 +11,6 @@ import type { LLM } from "@llamaindex/core/llms";
 import { AsyncLocalStorage, getEnv } from "@llamaindex/env";
 import type { ServiceContext } from "./ServiceContext.js";
 import type { BaseEmbedding } from "./embeddings/types.js";
-import {
-  getCallbackManager,
-  setCallbackManager,
-  withCallbackManager,
-} from "./internal/settings/CallbackManager.js";
 import {
   getEmbeddedModel,
   setEmbeddedModel,
@@ -129,18 +126,18 @@ class GlobalSettings implements Config {
   }
 
   get callbackManager(): CallbackManager {
-    return getCallbackManager();
+    return CoreSettings.callbackManager;
   }
 
   set callbackManager(callbackManager: CallbackManager) {
-    setCallbackManager(callbackManager);
+    CoreSettings.callbackManager = callbackManager;
   }
 
   withCallbackManager<Result>(
     callbackManager: CallbackManager,
     fn: () => Result,
   ): Result {
-    return withCallbackManager(callbackManager, fn);
+    return CoreSettings.withCallbackManager(callbackManager, fn);
   }
 
   set chunkSize(chunkSize: number | undefined) {

packages/llamaindex/src/agent/base.ts

@@ -5,6 +5,7 @@ import type {
   MessageContent,
   ToolOutput,
 } from "@llamaindex/core/llms";
+import { wrapEventCaller } from "@llamaindex/core/utils";
 import { ReadableStream, TransformStream, randomUUID } from "@llamaindex/env";
 import { ChatHistory } from "../ChatHistory.js";
 import { EngineResponse } from "../EngineResponse.js";
@@ -14,9 +15,7 @@ import {
   type ChatEngineParamsNonStreaming,
   type ChatEngineParamsStreaming,
 } from "../engines/chat/index.js";
-import { wrapEventCaller } from "../internal/context/EventCaller.js";
 import { consoleLogger, emptyLogger } from "../internal/logger.js";
-import { getCallbackManager } from "../internal/settings/CallbackManager.js";
 import { isAsyncIterable } from "../internal/utils.js";
 import { ObjectRetriever } from "../objects/index.js";
 import type {
@@ -69,10 +68,8 @@ export function createTaskOutputStream<
         taskOutputs.push(output);
         controller.enqueue(output);
       };
-      getCallbackManager().dispatchEvent("agent-start", {
-        payload: {
-          startStep: step,
-        },
+      Settings.callbackManager.dispatchEvent("agent-start", {
+        startStep: step,
       });
 
       context.logger.log("Starting step(id, %s).", step.id);
@@ -93,10 +90,8 @@ export function createTaskOutputStream<
           "Final step(id, %s) reached, closing task.",
           step.id,
         );
-        getCallbackManager().dispatchEvent("agent-end", {
-          payload: {
-            endStep: step,
-          },
+        Settings.callbackManager.dispatchEvent("agent-end", {
+          endStep: step,
         });
         controller.close();
       }

packages/llamaindex/src/agent/react.ts

@@ -1,3 +1,4 @@
+import type { JSONObject, JSONValue } from "@llamaindex/core/global";
 import type {
   BaseTool,
   ChatMessage,
@@ -5,15 +6,14 @@ import type {
   ChatResponseChunk,
   LLM,
 } from "@llamaindex/core/llms";
+import { extractText } from "@llamaindex/core/utils";
 import { randomUUID, ReadableStream } from "@llamaindex/env";
 import { getReACTAgentSystemHeader } from "../internal/prompt/react.js";
 import {
   isAsyncIterable,
   stringifyJSONToMessageContent,
 } from "../internal/utils.js";
-import { extractText } from "../llm/utils.js";
 import { Settings } from "../Settings.js";
-import type { JSONObject, JSONValue } from "../types.js";
 import { AgentRunner, AgentWorker, type AgentParamsBase } from "./base.js";
 import type { TaskHandler } from "./types.js";
 import {
@@ -66,7 +66,7 @@ function reasonFormatter(reason: Reason): string | Promise<string> {
 }
 
 function extractJsonStr(text: string): string {
-  const pattern = /\{.*}/s;
+  const pattern = /\{.*\}/s;
   const match = text.match(pattern);
 
   if (!match) {
@@ -98,7 +98,7 @@ function extractToolUse(
   inputText: string,
 ): [thought: string, action: string, input: string] {
   const pattern =
-    /\s*Thought: (.*?)\nAction: ([a-zA-Z0-9_]+).*?\.*Input: .*?(\{.*?})/s;
+    /\s*Thought: (.*?)\nAction: ([a-zA-Z0-9_]+).*?\.*Input: .*?(\{.*?\})/s;
 
   const match = inputText.match(pattern);
 

packages/llamaindex/src/agent/types.ts

@@ -9,7 +9,6 @@ import type {
 } from "@llamaindex/core/llms";
 import { ReadableStream } from "@llamaindex/env";
 import type { Logger } from "../internal/logger.js";
-import type { BaseEvent } from "../internal/type.js";
 import type { UUID } from "../types.js";
 
 export type AgentTaskContext<
@@ -90,9 +89,9 @@ export type TaskHandler<
   ) => void,
 ) => Promise<void>;
 
-export type AgentStartEvent = BaseEvent<{
+export type AgentStartEvent = {
   startStep: TaskStep;
-}>;
-export type AgentEndEvent = BaseEvent<{
+};
+export type AgentEndEvent = {
   endStep: TaskStep;
-}>;
+};

packages/llamaindex/src/agent/utils.ts

@@ -1,3 +1,8 @@
+import {
+  type JSONObject,
+  type JSONValue,
+  Settings,
+} from "@llamaindex/core/global";
 import type {
   BaseTool,
   ChatMessage,
@@ -14,13 +19,11 @@ import { baseToolWithCallSchema } from "@llamaindex/core/schema";
 import { ReadableStream } from "@llamaindex/env";
 import { z } from "zod";
 import type { Logger } from "../internal/logger.js";
-import { getCallbackManager } from "../internal/settings/CallbackManager.js";
 import {
   isAsyncIterable,
   prettifyError,
   stringifyJSONToMessageContent,
 } from "../internal/utils.js";
-import type { JSONObject, JSONValue } from "../types.js";
 import type { AgentParamsBase } from "./base.js";
 import type { TaskHandler } from "./types.js";
 
@@ -221,10 +224,8 @@ export async function callTool(
     };
   }
   try {
-    getCallbackManager().dispatchEvent("llm-tool-call", {
-      payload: {
-        toolCall: { ...toolCall, input },
-      },
+    Settings.callbackManager.dispatchEvent("llm-tool-call", {
+      toolCall: { ...toolCall, input },
     });
     output = await call.call(tool, input);
     logger.log(
@@ -237,11 +238,9 @@ export async function callTool(
       output,
       isError: false,
     };
-    getCallbackManager().dispatchEvent("llm-tool-result", {
-      payload: {
-        toolCall: { ...toolCall, input },
-        toolResult: { ...toolOutput },
-      },
+    Settings.callbackManager.dispatchEvent("llm-tool-result", {
+      toolCall: { ...toolCall, input },
+      toolResult: { ...toolOutput },
     });
     return toolOutput;
   } catch (e) {

packages/llamaindex/src/callbacks/CallbackManager.ts

@@ -1,230 +0,0 @@
-import type { Anthropic } from "@anthropic-ai/sdk";
-import type { MessageContent } from "@llamaindex/core/llms";
-import type { NodeWithScore } from "@llamaindex/core/schema";
-import { CustomEvent } from "@llamaindex/env";
-import type { AgentEndEvent, AgentStartEvent } from "../agent/types.js";
-import {
-  EventCaller,
-  getEventCaller,
-} from "../internal/context/EventCaller.js";
-import type {
-  LLMEndEvent,
-  LLMStartEvent,
-  LLMStreamEvent,
-  LLMToolCallEvent,
-  LLMToolResultEvent,
-  RetrievalEndEvent,
-  RetrievalStartEvent,
-} from "../llm/types.js";
-
-export class LlamaIndexCustomEvent<T = any> extends CustomEvent<T> {
-  reason: EventCaller | null;
-  private constructor(
-    event: string,
-    options?: CustomEventInit & {
-      reason?: EventCaller | null;
-    },
-  ) {
-    super(event, options);
-    this.reason = options?.reason ?? null;
-  }
-
-  static fromEvent<Type extends keyof LlamaIndexEventMaps>(
-    type: Type,
-    detail: LlamaIndexEventMaps[Type]["detail"],
-  ) {
-    return new LlamaIndexCustomEvent(type, {
-      detail: detail,
-      reason: getEventCaller(),
-    });
-  }
-}
-
-/**
- * This type is used to define the event maps.
- */
-export interface LlamaIndexEventMaps {
-  /**
-   * @deprecated
-   */
-  retrieve: CustomEvent<RetrievalCallbackResponse>;
-  "retrieve-start": RetrievalStartEvent;
-  "retrieve-end": RetrievalEndEvent;
-  /**
-   * @deprecated
-   */
-  stream: CustomEvent<StreamCallbackResponse>;
-  // llm events
-  "llm-start": LLMStartEvent;
-  "llm-end": LLMEndEvent;
-  "llm-tool-call": LLMToolCallEvent;
-  "llm-tool-result": LLMToolResultEvent;
-  "llm-stream": LLMStreamEvent;
-  // agent events
-  "agent-start": AgentStartEvent;
-  "agent-end": AgentEndEvent;
-}
-
-//#region @deprecated remove in the next major version
-//Specify StreamToken per mainstream LLM
-export interface DefaultStreamToken {
-  id: string;
-  object: string;
-  created: number;
-  model: string;
-  choices: {
-    index: number;
-    delta: {
-      content?: string | null;
-      role?: "user" | "assistant" | "system" | "function" | "tool";
-    };
-    finish_reason: string | null;
-  }[];
-}
-
-//OpenAI stream token schema is the default.
-//Note: Anthropic and Replicate also use similar token schemas.
-export type OpenAIStreamToken = DefaultStreamToken;
-export type AnthropicStreamToken = Anthropic.Completion;
-//
-//Callback Responses
-//
-//TODO: Write Embedding Callbacks
-
-//StreamCallbackResponse should let practitioners implement callbacks out of the box...
-//When custom streaming LLMs are involved, people are expected to write their own StreamCallbackResponses
-export interface StreamCallbackResponse {
-  index: number;
-  isDone?: boolean;
-  token?: DefaultStreamToken;
-}
-
-export interface RetrievalCallbackResponse {
-  query: MessageContent;
-  nodes: NodeWithScore[];
-}
-
-interface CallbackManagerMethods {
-  /**
-   * onLLMStream is called when a token is streamed from the LLM. Defining this
-   * callback auto sets the stream = True flag on the openAI createChatCompletion request.
-   * @deprecated will be removed in the next major version
-   */
-  onLLMStream: (params: StreamCallbackResponse) => Promise<void> | void;
-  /**
-   * onRetrieve is called as soon as the retriever finishes fetching relevant nodes.
-   * This callback allows you to handle the retrieved nodes even if the synthesizer
-   * is still running.
-   * @deprecated will be removed in the next major version
-   */
-  onRetrieve: (params: RetrievalCallbackResponse) => Promise<void> | void;
-}
-//#endregion
-
-const noop: (...args: any[]) => any = () => void 0;
-
-type EventHandler<Event extends CustomEvent> = (
-  event: Event & {
-    reason: EventCaller | null;
-  },
-) => void;
-
-export class CallbackManager implements CallbackManagerMethods {
-  /**
-   * @deprecated will be removed in the next major version
-   */
-  get onLLMStream(): CallbackManagerMethods["onLLMStream"] {
-    return async (response) => {
-      await Promise.all(
-        this.#handlers
-          .get("stream")!
-          .map((handler) =>
-            handler(LlamaIndexCustomEvent.fromEvent("stream", response)),
-          ),
-      );
-    };
-  }
-
-  /**
-   * @deprecated will be removed in the next major version
-   */
-  get onRetrieve(): CallbackManagerMethods["onRetrieve"] {
-    return async (response) => {
-      await Promise.all(
-        this.#handlers
-          .get("retrieve")!
-          .map((handler) =>
-            handler(LlamaIndexCustomEvent.fromEvent("retrieve", response)),
-          ),
-      );
-    };
-  }
-
-  /**
-   * @deprecated will be removed in the next major version
-   */
-  set onLLMStream(_: never) {
-    throw new Error("onLLMStream is deprecated. Use on('stream') instead");
-  }
-
-  /**
-   * @deprecated will be removed in the next major version
-   */
-  set onRetrieve(_: never) {
-    throw new Error("onRetrieve is deprecated. Use `on('retrieve')` instead");
-  }
-
-  #handlers = new Map<keyof LlamaIndexEventMaps, EventHandler<CustomEvent>[]>();
-
-  constructor(handlers?: Partial<CallbackManagerMethods>) {
-    const onLLMStream = handlers?.onLLMStream ?? noop;
-    this.on("stream", (event) => onLLMStream(event.detail));
-    const onRetrieve = handlers?.onRetrieve ?? noop;
-    this.on("retrieve", (event) => onRetrieve(event.detail));
-  }
-
-  on<
-    K extends keyof LlamaIndexEventMaps,
-    H extends EventHandler<LlamaIndexEventMaps[K]>,
-  >(event: K, handler: H) {
-    if (!this.#handlers.has(event)) {
-      this.#handlers.set(event, []);
-    }
-    this.#handlers.get(event)!.push(handler);
-    return this;
-  }
-
-  off<
-    K extends keyof LlamaIndexEventMaps,
-    H extends EventHandler<LlamaIndexEventMaps[K]>,
-  >(event: K, handler: H) {
-    if (!this.#handlers.has(event)) {
-      return;
-    }
-    const handlers = this.#handlers.get(event)!;
-    const index = handlers.indexOf(handler);
-    if (index > -1) {
-      handlers.splice(index, 1);
-    }
-    return this;
-  }
-
-  dispatchEvent<K extends keyof LlamaIndexEventMaps>(
-    event: K,
-    detail: LlamaIndexEventMaps[K]["detail"],
-  ) {
-    const handlers = this.#handlers.get(event);
-    if (!handlers) {
-      return;
-    }
-    queueMicrotask(() => {
-      handlers.forEach((handler) =>
-        handler(
-          LlamaIndexCustomEvent.fromEvent(event, {
-            ...detail,
-          }),
-        ),
-      );
-    });
-  }
-}

packages/llamaindex/src/cloud/LlamaCloudRetriever.ts

@@ -4,11 +4,11 @@ import {
   type RetrievalParams,
   type TextNodeWithScore,
 } from "@llamaindex/cloud/api";
+import { Settings } from "@llamaindex/core/global";
 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 { wrapEventCaller } from "../internal/context/EventCaller.js";
-import { extractText } from "../llm/utils.js";
 import type { ClientParams, CloudConstructorParams } from "./constants.js";
 import { DEFAULT_PROJECT_NAME } from "./constants.js";
 import { initService } from "./utils.js";
@@ -28,9 +28,14 @@ export class LlamaCloudRetriever implements BaseRetriever {
     nodes: TextNodeWithScore[],
   ): NodeWithScore[] {
     return nodes.map((node: TextNodeWithScore) => {
+      const textNode = jsonToNode(node.node, ObjectType.TEXT);
+      textNode.metadata = {
+        ...textNode.metadata,
+        ...node.node.extra_info, // append LlamaCloud extra_info to node metadata (file_name, pipeline_id, etc.)
+      };
       return {
         // Currently LlamaCloud only supports text nodes
-        node: jsonToNode(node.node, ObjectType.TEXT),
+        node: textNode,
         score: node.score,
       };
     });
@@ -83,6 +88,13 @@ export class LlamaCloudRetriever implements BaseRetriever {
         },
       });
 
-    return this.resultNodesToNodeWithScore(results.retrieval_nodes);
+    const nodesWithScores = this.resultNodesToNodeWithScore(
+      results.retrieval_nodes,
+    );
+    Settings.callbackManager.dispatchEvent("retrieve-end", {
+      query,
+      nodes: nodesWithScores,
+    });
+    return nodesWithScores;
   }
 }

packages/llamaindex/src/embeddings/DeepInfraEmbedding.ts

@@ -1,6 +1,6 @@
 import type { MessageContentDetail } from "@llamaindex/core/llms";
+import { extractSingleText } from "@llamaindex/core/utils";
 import { getEnv } from "@llamaindex/env";
-import { extractSingleText } from "../llm/utils.js";
 import { BaseEmbedding } from "./types.js";
 
 const DEFAULT_MODEL = "sentence-transformers/clip-ViT-B-32";

packages/llamaindex/src/embeddings/MixedbreadAIEmbeddings.ts

@@ -0,0 +1,170 @@
+import { getEnv } from "@llamaindex/env";
+import { MixedbreadAI, MixedbreadAIClient } from "@mixedbread-ai/sdk";
+import { BaseEmbedding, type EmbeddingInfo } from "./types.js";
+
+type EmbeddingsRequestWithoutInput = Omit<
+  MixedbreadAI.EmbeddingsRequest,
+  "input"
+>;
+
+/**
+ * Interface extending EmbeddingsParams with additional
+ * parameters specific to the MixedbreadAIEmbeddings class.
+ */
+export interface MixedbreadAIEmbeddingsParams
+  extends Omit<EmbeddingsRequestWithoutInput, "model"> {
+  /**
+   * The model to use for generating embeddings.
+   * @default {"mixedbread-ai/mxbai-embed-large-v1"}
+   */
+  model?: string;
+
+  /**
+   * The API key to use.
+   * @default {process.env.MXBAI_API_KEY}
+   */
+  apiKey?: string;
+
+  /**
+   * The base URL for the API.
+   */
+  baseUrl?: string;
+
+  /**
+   * The maximum number of documents to embed in a single request.
+   * @default {128}
+   */
+  embedBatchSize?: number;
+
+  /**
+   * The embed info for the model.
+   */
+  embedInfo?: EmbeddingInfo;
+
+  /**
+   * The maximum number of retries to attempt.
+   * @default {3}
+   */
+  maxRetries?: number;
+
+  /**
+   * Timeouts for the request.
+   */
+  timeoutInSeconds?: number;
+}
+
+/**
+ * Class for generating embeddings using the mixedbread ai API.
+ *
+ * This class leverages the model "mixedbread-ai/mxbai-embed-large-v1" to generate
+ * embeddings for text documents. The embeddings can be used for various NLP tasks
+ * such as similarity comparison, clustering, or as features in machine learning models.
+ *
+ * @example
+ * const mxbai = new MixedbreadAIEmbeddings({ apiKey: 'your-api-key' });
+ * const texts = ["Baking bread is fun", "I love baking"];
+ * const result = await mxbai.getTextEmbeddings(texts);
+ * console.log(result);
+ *
+ * @example
+ * const mxbai = new MixedbreadAIEmbeddings({
+ *  apiKey: 'your-api-key',
+ *  model: 'mixedbread-ai/mxbai-embed-large-v1',
+ *  encodingFormat: MixedbreadAI.EncodingFormat.Binary,
+ *  dimensions: 512,
+ *  normalized: true,
+ * });
+ * const query = "Represent this sentence for searching relevant passages: Is baking bread fun?";
+ * const result = await mxbai.getTextEmbedding(query);
+ * console.log(result);
+ */
+export class MixedbreadAIEmbeddings extends BaseEmbedding {
+  requestParams: EmbeddingsRequestWithoutInput;
+  requestOptions: MixedbreadAIClient.RequestOptions;
+  private client: MixedbreadAIClient;
+
+  /**
+   * Constructor for MixedbreadAIEmbeddings.
+   * @param {Partial<MixedbreadAIEmbeddingsParams>} params - An optional object with properties to configure the instance.
+   * @throws {Error} If the API key is not provided or found in the environment variables.
+   * @throws {Error} If the batch size exceeds 256.
+   */
+  constructor(params?: Partial<MixedbreadAIEmbeddingsParams>) {
+    super();
+
+    const apiKey = params?.apiKey ?? getEnv("MXBAI_API_KEY");
+    if (!apiKey) {
+      throw new Error(
+        "mixedbread ai API key not found. Either provide it in the constructor or set the 'MXBAI_API_KEY' environment variable.",
+      );
+    }
+    if (params?.embedBatchSize && params?.embedBatchSize > 256) {
+      throw new Error(
+        "The maximum batch size for mixedbread ai embeddings API is 256.",
+      );
+    }
+
+    this.embedBatchSize = params?.embedBatchSize ?? 128;
+    this.embedInfo = params?.embedInfo;
+    this.requestParams = {
+      model: params?.model ?? "mixedbread-ai/mxbai-embed-large-v1",
+      normalized: params?.normalized,
+      dimensions: params?.dimensions,
+      encodingFormat: params?.encodingFormat,
+      truncationStrategy: params?.truncationStrategy,
+      prompt: params?.prompt,
+    };
+    this.requestOptions = {
+      timeoutInSeconds: params?.timeoutInSeconds,
+      maxRetries: params?.maxRetries ?? 3,
+      // Support for this already exists in the python sdk and will be added to the js sdk soon
+      // @ts-ignore
+      additionalHeaders: {
+        "user-agent": "@mixedbread-ai/llamaindex-ts-sdk",
+      },
+    };
+    this.client = new MixedbreadAIClient({
+      apiKey,
+      environment: params?.baseUrl,
+    });
+  }
+
+  /**
+   * Generates an embedding for a single text.
+   * @param {string} text - A string to generate an embedding for.
+   * @returns {Promise<number[]>} A Promise that resolves to an array of numbers representing the embedding.
+   *
+   * @example
+   * const query = "Represent this sentence for searching relevant passages: Is baking bread fun?";
+   * const result = await mxbai.getTextEmbedding(text);
+   * console.log(result);
+   */
+  async getTextEmbedding(text: string): Promise<number[]> {
+    return (await this.getTextEmbeddings([text]))[0];
+  }
+
+  /**
+   * Generates embeddings for an array of texts.
+   * @param {string[]} texts - An array of strings to generate embeddings for.
+   * @returns {Promise<Array<number[]>>} A Promise that resolves to an array of embeddings.
+   *
+   * @example
+   * const texts = ["Baking bread is fun", "I love baking"];
+   * const result = await mxbai.getTextEmbeddings(texts);
+   * console.log(result);
+   */
+  async getTextEmbeddings(texts: string[]): Promise<Array<number[]>> {
+    if (texts.length === 0) {
+      return [];
+    }
+
+    const response = await this.client.embeddings(
+      {
+        ...this.requestParams,
+        input: texts,
+      },
+      this.requestOptions,
+    );
+    return response.data.map((d) => d.embedding as number[]);
+  }
+}

packages/llamaindex/src/embeddings/MultiModalEmbedding.ts

@@ -7,7 +7,7 @@ import {
   type BaseNode,
   type ImageType,
 } from "@llamaindex/core/schema";
-import { extractImage, extractSingleText } from "../llm/utils.js";
+import { extractImage, extractSingleText } from "@llamaindex/core/utils";
 import { BaseEmbedding, batchEmbeddings } from "./types.js";
 
 /*

packages/llamaindex/src/embeddings/index.ts

@@ -4,6 +4,7 @@ export * from "./GeminiEmbedding.js";
 export { HuggingFaceInferenceAPIEmbedding } from "./HuggingFaceEmbedding.js";
 export * from "./JinaAIEmbedding.js";
 export * from "./MistralAIEmbedding.js";
+export * from "./MixedbreadAIEmbeddings.js";
 export * from "./MultiModalEmbedding.js";
 export { OllamaEmbedding } from "./OllamaEmbedding.js";
 export * from "./OpenAIEmbedding.js";

packages/llamaindex/src/embeddings/types.ts

@@ -1,9 +1,9 @@
 import type { MessageContentDetail } from "@llamaindex/core/llms";
 import type { BaseNode } from "@llamaindex/core/schema";
 import { MetadataMode } from "@llamaindex/core/schema";
+import { extractSingleText } from "@llamaindex/core/utils";
 import { type Tokenizers } from "@llamaindex/env";
 import type { TransformComponent } from "../ingestion/types.js";
-import { extractSingleText } from "../llm/utils.js";
 import { truncateMaxTokens } from "./tokenizer.js";
 import { SimilarityType, similarity } from "./utils.js";
 

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

@@ -1,4 +1,9 @@
 import type { ChatMessage, LLM } from "@llamaindex/core/llms";
+import {
+  extractText,
+  streamReducer,
+  wrapEventCaller,
+} from "@llamaindex/core/utils";
 import type { ChatHistory } from "../../ChatHistory.js";
 import { getHistory } from "../../ChatHistory.js";
 import type { EngineResponse } from "../../EngineResponse.js";
@@ -9,8 +14,6 @@ import {
 } from "../../Prompt.js";
 import type { ServiceContext } from "../../ServiceContext.js";
 import { llmFromSettingsOrContext } from "../../Settings.js";
-import { wrapEventCaller } from "../../internal/context/EventCaller.js";
-import { extractText, streamReducer } from "../../llm/utils.js";
 import { PromptMixin } from "../../prompts/index.js";
 import type { QueryEngine } from "../../types.js";
 import type {

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

@@ -4,18 +4,18 @@ import type {
   MessageContent,
   MessageType,
 } from "@llamaindex/core/llms";
+import {
+  extractText,
+  streamConverter,
+  streamReducer,
+  wrapEventCaller,
+} from "@llamaindex/core/utils";
 import type { ChatHistory } from "../../ChatHistory.js";
 import { getHistory } from "../../ChatHistory.js";
 import { EngineResponse } from "../../EngineResponse.js";
 import type { ContextSystemPrompt } from "../../Prompt.js";
 import type { BaseRetriever } from "../../Retriever.js";
 import { Settings } from "../../Settings.js";
-import { wrapEventCaller } from "../../internal/context/EventCaller.js";
-import {
-  extractText,
-  streamConverter,
-  streamReducer,
-} from "../../llm/utils.js";
 import type { BaseNodePostprocessor } from "../../postprocessors/index.js";
 import { PromptMixin } from "../../prompts/Mixin.js";
 import { DefaultContextGenerator } from "./DefaultContextGenerator.js";

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

@@ -1,10 +1,13 @@
 import type { LLM } from "@llamaindex/core/llms";
+import {
+  streamConverter,
+  streamReducer,
+  wrapEventCaller,
+} from "@llamaindex/core/utils";
 import type { ChatHistory } from "../../ChatHistory.js";
 import { getHistory } from "../../ChatHistory.js";
 import { EngineResponse } from "../../EngineResponse.js";
 import { Settings } from "../../Settings.js";
-import { wrapEventCaller } from "../../internal/context/EventCaller.js";
-import { streamConverter, streamReducer } from "../../llm/utils.js";
 import type {
   ChatEngine,
   ChatEngineParamsNonStreaming,

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

@@ -1,9 +1,9 @@
 import type { NodeWithScore } from "@llamaindex/core/schema";
+import { wrapEventCaller } from "@llamaindex/core/utils";
 import type { EngineResponse } from "../../EngineResponse.js";
-import type { BaseRetriever } from "../../Retriever.js";
-import { wrapEventCaller } from "../../internal/context/EventCaller.js";
 import type { BaseNodePostprocessor } from "../../postprocessors/index.js";
 import { PromptMixin } from "../../prompts/Mixin.js";
+import type { BaseRetriever } from "../../Retriever.js";
 import type { BaseSynthesizer } from "../../synthesizers/index.js";
 import { ResponseSynthesizer } from "../../synthesizers/index.js";
 import type {

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

@@ -17,7 +17,7 @@ import type {
 } from "../../types.js";
 
 import type { BaseTool, ToolMetadata } from "@llamaindex/core/llms";
-import { wrapEventCaller } from "../../internal/context/EventCaller.js";
+import { wrapEventCaller } from "@llamaindex/core/utils";
 import type { BaseQuestionGenerator, SubQuestion } from "./types.js";
 
 /**

packages/llamaindex/src/evaluation/Correctness.ts

@@ -1,6 +1,6 @@
 import type { ChatMessage, LLM } from "@llamaindex/core/llms";
 import { MetadataMode } from "@llamaindex/core/schema";
-import { extractText } from "../llm/utils.js";
+import { extractText } from "@llamaindex/core/utils";
 import { PromptMixin } from "../prompts/Mixin.js";
 import type { ServiceContext } from "../ServiceContext.js";
 import { llmFromSettingsOrContext } from "../Settings.js";

packages/llamaindex/src/index.edge.ts

@@ -1,7 +1,30 @@
+import type { AgentEndEvent, AgentStartEvent } from "./agent/types.js";
+import type { RetrievalEndEvent, RetrievalStartEvent } from "./llm/types.js";
+
+declare module "@llamaindex/core/global" {
+  export interface LlamaIndexEventMaps {
+    "retrieve-start": RetrievalStartEvent;
+    "retrieve-end": RetrievalEndEvent;
+    // agent events
+    "agent-start": AgentStartEvent;
+    "agent-end": AgentEndEvent;
+  }
+}
+
+export { CallbackManager } from "@llamaindex/core/global";
+export type {
+  JSONArray,
+  JSONObject,
+  JSONValue,
+  LLMEndEvent,
+  LLMStartEvent,
+  LLMStreamEvent,
+  LLMToolCallEvent,
+  LLMToolResultEvent,
+} from "@llamaindex/core/global";
 export * from "@llamaindex/core/llms";
 export * from "@llamaindex/core/schema";
 export * from "./agent/index.js";
-export * from "./callbacks/CallbackManager.js";
 export * from "./ChatHistory.js";
 export * from "./cloud/index.js";
 export * from "./constants.js";

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

@@ -32,8 +32,8 @@ import {
 } from "./utils.js";
 
 import type { LLM } from "@llamaindex/core/llms";
+import { extractText } from "@llamaindex/core/utils";
 import { llmFromSettingsOrContext } from "../../Settings.js";
-import { extractText } from "../../llm/utils.js";
 
 export interface KeywordIndexOptions {
   nodes?: BaseNode[];

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

@@ -3,19 +3,17 @@ import type {
   Document,
   NodeWithScore,
 } from "@llamaindex/core/schema";
+import { extractText, wrapEventCaller } from "@llamaindex/core/utils";
 import _ from "lodash";
 import type { ChoiceSelectPrompt } from "../../Prompt.js";
 import { defaultChoiceSelectPrompt } from "../../Prompt.js";
 import type { BaseRetriever, RetrieveParams } from "../../Retriever.js";
 import type { ServiceContext } from "../../ServiceContext.js";
 import {
-  Settings,
   llmFromSettingsOrContext,
   nodeParserFromSettingsOrContext,
 } from "../../Settings.js";
 import { RetrieverQueryEngine } from "../../engines/query/index.js";
-import { wrapEventCaller } from "../../internal/context/EventCaller.js";
-import { extractText } from "../../llm/utils.js";
 import type { BaseNodePostprocessor } from "../../postprocessors/index.js";
 import type { StorageContext } from "../../storage/StorageContext.js";
 import { storageContextFromDefaults } from "../../storage/StorageContext.js";
@@ -296,17 +294,10 @@ export class SummaryIndexRetriever implements BaseRetriever {
   async retrieve({ query }: RetrieveParams): Promise<NodeWithScore[]> {
     const nodeIds = this.index.indexStruct.nodes;
     const nodes = await this.index.docStore.getNodes(nodeIds);
-    const result = nodes.map((node) => ({
+    return nodes.map((node) => ({
       node: node,
       score: 1,
     }));
-
-    Settings.callbackManager.dispatchEvent("retrieve", {
-      query,
-      nodes: result,
-    });
-
-    return result;
   }
 }
 
@@ -376,11 +367,6 @@ export class SummaryIndexLLMRetriever implements BaseRetriever {
       results.push(...nodeWithScores);
     }
 
-    Settings.callbackManager.dispatchEvent("retrieve", {
-      query,
-      nodes: results,
-    });
-
     return results;
   }
 }