Release 0.9.9 (#1713)

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: marcusschiesser <17126+marcusschiesser@users.noreply.github.com>

CONTRIBUTING.md

@@ -41,8 +41,15 @@ pnpm install
 
 ### Build the packages
 
+You'll need Turbo to build the packages. If you don't have it, you can run it with `pnpx`.
+
+To build all packages, run:
+
 ```shell
 # Build all packages
+pnpx turbo build --filter "./packages/*"
+
+# Or if you have turbo installed, you can run:
 turbo build --filter "./packages/*"
 ```
 

apps/next/CHANGELOG.md

@@ -1,5 +1,79 @@
 # @llamaindex/doc
 
+## 0.1.9
+
+### Patch Changes
+
+- 4bac71d: Support binding additional argument to function tool
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+  - @llamaindex/cloud@3.0.8
+  - llamaindex@0.9.9
+  - @llamaindex/node-parser@1.0.7
+  - @llamaindex/openai@0.1.59
+  - @llamaindex/readers@2.0.7
+  - @llamaindex/workflow@0.0.15
+
+## 0.1.8
+
+### Patch Changes
+
+- Updated dependencies [4b49428]
+- Updated dependencies [bbc8c87]
+  - @llamaindex/workflow@0.0.14
+  - llamaindex@0.9.8
+
+## 0.1.7
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/core@0.5.6
+  - llamaindex@0.9.7
+  - @llamaindex/cloud@3.0.7
+  - @llamaindex/node-parser@1.0.6
+  - @llamaindex/openai@0.1.58
+  - @llamaindex/readers@2.0.6
+  - @llamaindex/workflow@0.0.13
+
+## 0.1.6
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+  - @llamaindex/workflow@0.0.12
+  - @llamaindex/cloud@3.0.6
+  - llamaindex@0.9.6
+  - @llamaindex/node-parser@1.0.5
+  - @llamaindex/openai@0.1.57
+  - @llamaindex/readers@2.0.5
+
+## 0.1.5
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+  - @llamaindex/cloud@3.0.5
+  - llamaindex@0.9.5
+  - @llamaindex/node-parser@1.0.4
+  - @llamaindex/openai@0.1.56
+  - @llamaindex/readers@2.0.4
+
+## 0.1.4
+
+### Patch Changes
+
+- Updated dependencies [cb256f2]
+- Updated dependencies [cb021e7]
+  - @llamaindex/openai@0.1.55
+  - @llamaindex/core@0.5.3
+  - llamaindex@0.9.4
+  - @llamaindex/cloud@3.0.4
+  - @llamaindex/node-parser@1.0.3
+  - @llamaindex/readers@2.0.3
+
 ## 0.1.3
 
 ### Patch Changes

apps/next/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/doc",
-  "version": "0.1.3",
+  "version": "0.1.9",
   "private": true,
   "scripts": {
     "build": "pnpm run build:docs && next build",

apps/next/src/content/docs/llamaindex/getting_started/setup/node.mdx

@@ -3,6 +3,8 @@ title: With Node.js/Bun/Deno
 description: In this guide, you'll learn how to use LlamaIndex with Node.js, Bun, and Deno.
 ---
 
+import { Tab, Tabs } from "fumadocs-ui/components/tabs";
+
 ## Adding environment variables
 
 By default, LlamaIndex uses OpenAI provider, which requires an API key. You can set the `OPENAI_API_KEY` environment variable to authenticate with OpenAI.
@@ -22,6 +24,26 @@ node --env-file .env your-script.js
 
 For more information, see the [How to read environment variables from Node.js](https://nodejs.org/en/learn/command-line/how-to-read-environment-variables-from-nodejs).
 
+## Performance Optimization
+
+By the default, we are using `js-tiktoken` for tokenization. You can install `gpt-tokenizer` which is then automatically used by LlamaIndex to get a 60x speedup for tokenization:
+
+<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
+	```shell tab="npm"
+	npm install gpt-tokenizer
+	```
+
+	```shell tab="yarn"
+	yarn add gpt-tokenizer
+	```
+
+	```shell tab="pnpm"
+	pnpm add gpt-tokenizer
+	```
+</Tabs>
+
+> Note: This only works for Node.js
+
 ## TypeScript support
 
 <Card

apps/next/src/content/docs/llamaindex/modules/agent_workflow.mdx

@@ -0,0 +1,139 @@
+---
+title: Agent Workflow
+---
+
+import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
+import CodeSource from "!raw-loader!../../../../../../../examples/agentworkflow/blog_writer.ts";
+import { Tab, Tabs } from "fumadocs-ui/components/tabs";
+
+`AgentWorkflow` is a powerful system that enables you to create and orchestrate one or multiple agents with tools to perform specific tasks. It's built on top of the base `Workflow` system and provides a streamlined interface for agent interactions.
+
+## Installation
+
+You'll need to install the `@llamaindex/workflow` package:
+
+<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
+	```shell tab="npm"
+	npm install @llamaindex/workflow
+	```
+
+	```shell tab="yarn"
+	yarn add @llamaindex/workflow
+	```
+
+	```shell tab="pnpm"
+	pnpm add @llamaindex/workflow
+	```
+</Tabs>
+
+## Usage
+
+### Single Agent Workflow
+
+The simplest use case is creating a single agent with specific tools. Here's an example of creating an assistant that tells jokes:
+
+```typescript
+import { AgentWorkflow, FunctionTool } from "llamaindex";
+import { OpenAI } from "@llamaindex/openai";
+
+// Define a joke-telling tool
+const jokeTool = FunctionTool.from(
+  () => "Baby Llama is called cria",
+  {
+    name: "joke",
+    description: "Use this tool to get a joke",
+  }
+);
+
+// Create an agent workflow with the tool
+const workflow = AgentWorkflow.fromTools({
+  tools: [jokeTool],
+  llm: new OpenAI({
+    model: "gpt-4o-mini",
+  }),
+});
+
+// Run the workflow
+const result = await workflow.run("Tell me something funny");
+console.log(result); // Baby Llama is called cria
+```
+
+### Event Streaming
+
+`AgentWorkflow` provides a unified interface for event streaming, making it easy to track and respond to different events during execution:
+
+```typescript
+import { AgentToolCall, AgentStream } from "llamaindex";
+
+// Get the workflow execution context
+const context = workflow.run("Tell me something funny");
+
+// Stream and handle events
+for await (const event of context) {
+  if (event instanceof AgentToolCall) {
+    console.log(`Tool being called: ${event.data.toolName}`);
+  }
+  if (event instanceof AgentStream) {
+    process.stdout.write(event.data.delta);
+  }
+}
+```
+
+### Multi-Agent Workflow
+
+`AgentWorkflow` can orchestrate multiple agents, enabling complex interactions and task handoffs. Each agent in a multi-agent workflow requires:
+
+- `name`: Unique identifier for the agent
+- `description`: Purpose description used for task routing
+- `tools`: Array of tools the agent can use
+- `canHandoffTo` (optional): Array of agent names or agent instances that this agent can delegate tasks to
+
+Here's an example of a multi-agent system that combines joke-telling and weather information:
+
+```typescript
+import { AgentWorkflow, FunctionAgent, FunctionTool } from "llamaindex";
+import { OpenAI } from "@llamaindex/openai";
+import { z } from "zod";
+
+// Create a weather agent
+const weatherAgent = new FunctionAgent({
+  name: "WeatherAgent",
+  description: "Provides weather information for any city",
+  tools: [
+    FunctionTool.from(
+      ({ city }: { city: string }) => `The weather in ${city} is sunny`,
+      {
+        name: "fetchWeather",
+        description: "Get weather information for a city",
+        parameters: z.object({
+          city: z.string(),
+        }),
+      }
+    ),
+  ],
+  llm: new OpenAI({ model: "gpt-4o-mini" }),
+});
+
+// Create a joke-telling agent
+const jokeAgent = new FunctionAgent({
+  name: "JokeAgent",
+  description: "Tells jokes and funny stories",
+  tools: [jokeTool], // Using the joke tool defined earlier
+  llm: new OpenAI({ model: "gpt-4o-mini" }),
+  canHandoffTo: [weatherAgent], // Can hand off to the weather agent
+});
+
+// Create the multi-agent workflow
+const workflow = new AgentWorkflow({
+  agents: [jokeAgent, weatherAgent],
+  rootAgent: jokeAgent, // Start with the joke agent
+});
+
+// Run the workflow
+const result = await workflow.run(
+  "Give me a morning greeting with a joke and the weather in San Francisco"
+);
+```
+
+The workflow will coordinate between agents, allowing them to handle different aspects of the request and hand off tasks when appropriate.
+

apps/next/src/content/docs/llamaindex/modules/embeddings/available_embeddings/jinaai.mdx

@@ -2,10 +2,11 @@
 title: Jina AI
 ---
 
-To use Jina AI embeddings, you need to import `JinaAIEmbedding` from `llamaindex`.
+To use Jina AI embeddings, you need to import `JinaAIEmbedding` from `@llamaindex/jinaai`.
 
 ```ts
-import { JinaAIEmbedding, Settings } from "llamaindex";
+import { Settings } from "llamaindex";
+import { JinaAIEmbedding } from "@llamaindex/jinaai";
 
 Settings.embedModel = new JinaAIEmbedding();
 

apps/next/src/content/docs/llamaindex/modules/embeddings/available_embeddings/together.mdx

@@ -2,10 +2,11 @@
 title: Together
 ---
 
-To use together embeddings, you need to import `TogetherEmbedding` from `llamaindex`.
+To use together embeddings, you need to import `TogetherEmbedding` from `@llamaindex/together`.
 
 ```ts
-import { TogetherEmbedding, Settings } from "llamaindex";
+import { Settings } from "llamaindex";
+import { TogetherEmbedding } from "@llamaindex/together";
 
 Settings.embedModel = new TogetherEmbedding({
   apiKey: "<YOUR_API_KEY>",

apps/next/src/content/docs/llamaindex/modules/embeddings/available_embeddings/voyageai.mdx

@@ -0,0 +1,46 @@
+---
+title: VoyageAI
+---
+
+To use VoyageAI embeddings, you need to import `VoyageAIEmbedding` from `@llamaindex/voyage-ai`.
+
+## Installation
+
+import { Tab, Tabs } from "fumadocs-ui/components/tabs";
+
+<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
+	```shell tab="npm"
+	npm install llamaindex @llamaindex/voyage-ai
+	```
+
+	```shell tab="yarn"
+	yarn add llamaindex @llamaindex/voyage-ai
+	```
+
+	```shell tab="pnpm"
+	pnpm add llamaindex @llamaindex/voyage-ai
+	```
+</Tabs>
+
+```ts
+import { VoyageAIEmbedding } from "@llamaindex/voyage-ai";
+import { Document, Settings, VectorStoreIndex } from "llamaindex";
+
+Settings.embedModel = new VoyageAIEmbedding();
+
+const document = new Document({ text: essay, id_: "essay" });
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+
+const queryEngine = index.asQueryEngine();
+
+const query = "What is the meaning of life?";
+
+const results = await queryEngine.query({
+  query,
+});
+```
+
+## API Reference
+
+- [VoyageAIEmbedding](/docs/api/classes/VoyageAIEmbedding)

apps/next/src/content/docs/llamaindex/modules/embeddings/index.mdx

@@ -37,6 +37,31 @@ Settings.embedModel = new OpenAIEmbedding({
 
 For local embeddings, you can use the [HuggingFace](/docs/llamaindex/modules/embeddings/available_embeddings/huggingface) embedding model.
 
+## Local Ollama Embeddings With Remote Host
+
+Ollama provides a way to run embedding models locally or connect to a remote Ollama instance. This is particularly useful when you need to:
+- Run embeddings without relying on external API services
+- Use custom embedding models
+- Connect to a shared Ollama instance in your network
+
+The ENV variable method you will find elsewhere sometimes may not work with the OllamaEmbedding class. Also note, you'll need to change the host
+in the Ollama server to `0.0.0.0` to allow connections from other machines.
+
+To use Ollama embeddings with a remote host, you need to specify the host URL in the configuration like this:
+
+```typescript
+import { OllamaEmbedding } from "@llamaindex/ollama";
+import { Settings } from "llamaindex";
+
+// Configure Ollama with a remote host
+Settings.embedModel = new OllamaEmbedding({
+  model: "nomic-embed-text",
+  config: {
+    host: "http://your-ollama-host:11434"
+  }
+});
+```
+
 ## Available Embeddings
 
 Most available embeddings are listed in the sidebar on the left.

apps/next/src/content/docs/llamaindex/modules/llms/available_llms/deepseek.mdx

@@ -7,7 +7,8 @@ title: DeepSeek LLM
 ## Usage
 
 ```ts
-import { DeepSeekLLM, Settings } from "llamaindex";
+import { Settings } from "llamaindex";
+import { DeepSeekLLM } from "@llamaindex/deepseek";
 
 Settings.llm = new DeepSeekLLM({
   apiKey: "<YOUR_API_KEY>",
@@ -18,7 +19,8 @@ Settings.llm = new DeepSeekLLM({
 ## Example
 
 ```ts
-import { DeepSeekLLM, Document, VectorStoreIndex, Settings } from "llamaindex";
+import { Document, VectorStoreIndex, Settings } from "llamaindex";
+import { DeepSeekLLM } from "@llamaindex/deepseek";
 
 const deepseekLlm = new DeepSeekLLM({
   apiKey: "<YOUR_API_KEY>",

apps/next/src/content/docs/llamaindex/modules/llms/available_llms/fireworks.mdx

@@ -7,7 +7,8 @@ title: Fireworks LLM
 ## Usage
 
 ```ts
-import { FireworksLLM, Settings } from "llamaindex";
+import { Settings } from "llamaindex";
+import { FireworksLLM } from "@llamaindex/fireworks";
 
 Settings.llm = new FireworksLLM({
   apiKey: "<YOUR_API_KEY>",

apps/next/src/content/docs/llamaindex/modules/llms/available_llms/together.mdx

@@ -23,7 +23,8 @@ import { Tab, Tabs } from "fumadocs-ui/components/tabs";
 ## Usage
 
 ```ts
-import { Settings, TogetherLLM } from "llamaindex";
+import { Settings } from "llamaindex";
+import { TogetherLLM } from "@llamaindex/together";
 
 Settings.llm = new TogetherLLM({
   apiKey: "<YOUR_API_KEY>",

apps/next/src/content/docs/llamaindex/modules/tool/index.mdx

@@ -0,0 +1,60 @@
+---
+title: Tools
+---
+
+A "tool" is a utility that can be called by an agent on behalf of an LLM.
+A tool can be called to perform custom actions, or retrieve extra information based on the LLM-generated input.
+A result from a tool call can be used by subsequent steps in a workflow, or to compute a final answer.
+For example, a "weather tool" could fetch some live weather information from a geographical location.
+
+## Function tool
+
+Function tools are implemented with the `FunctionTool` class.
+A `FunctionTool` is constructed from a function with signature
+```ts
+(input: T, additionalArg?: AdditionalToolArgument) => R
+```
+where
+- `input` is generated by the LLM, `T` is the type defined by the tool `parameters`
+- `additionalArg` is an optional extra argument, see "Binding" below
+- `R` is the return type
+
+### Binding
+
+An additional argument can be bound to a tool, each tool call will be passed
+- the input provided by the LLM
+- the additional argument (extends object)
+
+Note: calling the `bind` method will return a new `FunctionTool` instance, without modifying the tool which `bind` is called on.
+
+Example to pass a `userToken` as additional argument:
+```ts
+// first arg is LLM input, second is bound arg
+const queryKnowledgeBase = async ({ question }, { userToken }) => {
+  const response = await fetch(`https://knowledge-base.com?token=${userToken}&query=${question}`);
+  // ...
+};
+
+// define tool as usual
+const kbTool = FunctionTool.from(queryKnowledgeBase, {
+  name: 'queryKnowledgeBase',
+  description: 'Query knowledge base',
+  parameters: {
+    type: 'object',
+    properties: {
+      question: {
+        type: 'string',
+        description: 'The user question',
+      },
+    },
+    required: ['question'],
+  },
+});
+
+// create an agent
+const additionalArg = { userToken: 'abcd1234' };
+const kbAgent = new LLMAgent({
+  tools: [kbTool.bind(additionalArg)],
+  // llm, systemPrompt etc
+})
+```

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

@@ -1,5 +1,44 @@
 # @llamaindex/cloudflare-worker-agent-test
 
+## 0.0.143
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 0.0.142
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.0.141
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+
+## 0.0.140
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 0.0.139
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 0.0.138
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 0.0.137
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.137",
+  "version": "0.0.143",
   "type": "module",
   "private": true,
   "scripts": {

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

@@ -1,5 +1,35 @@
 # @llamaindex/llama-parse-browser-test
 
+## 0.0.53
+
+### Patch Changes
+
+- @llamaindex/cloud@3.0.8
+
+## 0.0.52
+
+### Patch Changes
+
+- @llamaindex/cloud@3.0.7
+
+## 0.0.51
+
+### Patch Changes
+
+- @llamaindex/cloud@3.0.6
+
+## 0.0.50
+
+### Patch Changes
+
+- @llamaindex/cloud@3.0.5
+
+## 0.0.49
+
+### Patch Changes
+
+- @llamaindex/cloud@3.0.4
+
 ## 0.0.48
 
 ### Patch Changes

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

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/llama-parse-browser-test",
   "private": true,
-  "version": "0.0.48",
+  "version": "0.0.53",
   "type": "module",
   "scripts": {
     "dev": "vite",

e2e/examples/nextjs-agent/CHANGELOG.md

@@ -1,5 +1,44 @@
 # @llamaindex/next-agent-test
 
+## 0.1.143
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 0.1.142
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.1.141
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+
+## 0.1.140
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 0.1.139
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 0.1.138
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 0.1.137
 
 ### Patch Changes

e2e/examples/nextjs-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-agent-test",
-  "version": "0.1.137",
+  "version": "0.1.143",
   "private": true,
   "scripts": {
     "dev": "next dev",

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

@@ -1,5 +1,44 @@
 # test-edge-runtime
 
+## 0.1.142
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 0.1.141
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.1.140
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+
+## 0.1.139
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 0.1.138
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 0.1.137
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 0.1.136
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.136",
+  "version": "0.1.142",
   "private": true,
   "scripts": {
     "dev": "next dev",

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

@@ -1,5 +1,54 @@
 # @llamaindex/next-node-runtime
 
+## 0.1.9
+
+### Patch Changes
+
+- llamaindex@0.9.9
+- @llamaindex/huggingface@0.0.43
+- @llamaindex/readers@2.0.7
+
+## 0.1.8
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.1.7
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+  - @llamaindex/huggingface@0.0.42
+  - @llamaindex/readers@2.0.6
+
+## 0.1.6
+
+### Patch Changes
+
+- llamaindex@0.9.6
+- @llamaindex/huggingface@0.0.41
+- @llamaindex/readers@2.0.5
+
+## 0.1.5
+
+### Patch Changes
+
+- llamaindex@0.9.5
+- @llamaindex/huggingface@0.0.40
+- @llamaindex/readers@2.0.4
+
+## 0.1.4
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+  - @llamaindex/huggingface@0.0.39
+  - @llamaindex/readers@2.0.3
+
 ## 0.1.3
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.1.3",
+  "version": "0.1.9",
   "private": true,
   "scripts": {
     "dev": "next dev",

e2e/examples/vite-import-llamaindex/CHANGELOG.md

@@ -1,5 +1,44 @@
 # vite-import-llamaindex
 
+## 0.0.9
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 0.0.8
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.0.7
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+
+## 0.0.6
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 0.0.5
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 0.0.4
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 0.0.3
 
 ### Patch Changes

e2e/examples/vite-import-llamaindex/package.json

@@ -1,7 +1,7 @@
 {
   "name": "vite-import-llamaindex",
   "private": true,
-  "version": "0.0.3",
+  "version": "0.0.9",
   "type": "module",
   "scripts": {
     "build": "vite build",

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

@@ -1,5 +1,45 @@
 # @llamaindex/waku-query-engine-test
 
+## 0.0.143
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 0.0.142
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.0.141
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - llamaindex@0.9.7
+
+## 0.0.140
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 0.0.139
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 0.0.138
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 0.0.137
 
 ### Patch Changes

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

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.137",
+  "version": "0.0.143",
   "type": "module",
   "private": true,
   "scripts": {

examples/CHANGELOG.md

@@ -1,5 +1,223 @@
 # examples
 
+## 0.2.8
+
+### Patch Changes
+
+- Updated dependencies [58b3ee5]
+- Updated dependencies [4bac71d]
+- Updated dependencies [8bf1ca1]
+  - @llamaindex/google@0.1.0
+  - @llamaindex/core@0.5.7
+  - @llamaindex/anthropic@0.2.5
+  - @llamaindex/cloud@3.0.8
+  - llamaindex@0.9.9
+  - @llamaindex/node-parser@1.0.7
+  - @llamaindex/clip@0.0.43
+  - @llamaindex/cohere@0.0.12
+  - @llamaindex/deepinfra@0.0.43
+  - @llamaindex/huggingface@0.0.43
+  - @llamaindex/jinaai@0.0.3
+  - @llamaindex/mistral@0.0.12
+  - @llamaindex/mixedbread@0.0.12
+  - @llamaindex/ollama@0.0.47
+  - @llamaindex/openai@0.1.59
+  - @llamaindex/portkey-ai@0.0.40
+  - @llamaindex/replicate@0.0.40
+  - @llamaindex/astra@0.0.12
+  - @llamaindex/azure@0.1.7
+  - @llamaindex/chroma@0.0.12
+  - @llamaindex/firestore@1.0.5
+  - @llamaindex/milvus@0.1.7
+  - @llamaindex/mongodb@0.0.12
+  - @llamaindex/pinecone@0.0.12
+  - @llamaindex/postgres@0.0.40
+  - @llamaindex/qdrant@0.1.7
+  - @llamaindex/upstash@0.0.12
+  - @llamaindex/weaviate@0.0.12
+  - @llamaindex/vercel@0.0.18
+  - @llamaindex/voyage-ai@1.0.4
+  - @llamaindex/readers@2.0.7
+  - @llamaindex/workflow@0.0.15
+  - @llamaindex/deepseek@0.0.3
+  - @llamaindex/fireworks@0.0.3
+  - @llamaindex/groq@0.0.58
+  - @llamaindex/together@0.0.3
+  - @llamaindex/vllm@0.0.29
+
+## 0.2.7
+
+### Patch Changes
+
+- Updated dependencies [4b49428]
+- Updated dependencies [bbc8c87]
+- Updated dependencies [7ee4968]
+  - @llamaindex/workflow@0.0.14
+  - llamaindex@0.9.8
+  - @llamaindex/deepseek@0.0.2
+  - @llamaindex/fireworks@0.0.2
+  - @llamaindex/together@0.0.2
+  - @llamaindex/jinaai@0.0.2
+  - @llamaindex/google@0.0.14
+
+## 0.2.6
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+  - llamaindex@0.9.7
+  - @llamaindex/cloud@3.0.7
+  - @llamaindex/node-parser@1.0.6
+  - @llamaindex/anthropic@0.2.4
+  - @llamaindex/clip@0.0.42
+  - @llamaindex/cohere@0.0.11
+  - @llamaindex/deepinfra@0.0.42
+  - @llamaindex/google@0.0.13
+  - @llamaindex/groq@0.0.57
+  - @llamaindex/huggingface@0.0.42
+  - @llamaindex/mistral@0.0.11
+  - @llamaindex/mixedbread@0.0.11
+  - @llamaindex/ollama@0.0.46
+  - @llamaindex/openai@0.1.58
+  - @llamaindex/portkey-ai@0.0.39
+  - @llamaindex/replicate@0.0.39
+  - @llamaindex/astra@0.0.11
+  - @llamaindex/azure@0.1.6
+  - @llamaindex/chroma@0.0.11
+  - @llamaindex/firestore@1.0.4
+  - @llamaindex/milvus@0.1.6
+  - @llamaindex/mongodb@0.0.11
+  - @llamaindex/pinecone@0.0.11
+  - @llamaindex/postgres@0.0.39
+  - @llamaindex/qdrant@0.1.6
+  - @llamaindex/upstash@0.0.11
+  - @llamaindex/weaviate@0.0.11
+  - @llamaindex/voyage-ai@1.0.3
+  - @llamaindex/readers@2.0.6
+  - @llamaindex/workflow@0.0.13
+  - @llamaindex/vercel@0.0.17
+  - @llamaindex/vllm@0.0.28
+
+## 0.2.5
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+- Updated dependencies [fd74ba4]
+  - @llamaindex/core@0.5.5
+  - @llamaindex/workflow@0.0.12
+  - @llamaindex/voyage-ai@1.0.2
+  - @llamaindex/cloud@3.0.6
+  - llamaindex@0.9.6
+  - @llamaindex/node-parser@1.0.5
+  - @llamaindex/anthropic@0.2.3
+  - @llamaindex/clip@0.0.41
+  - @llamaindex/cohere@0.0.10
+  - @llamaindex/deepinfra@0.0.41
+  - @llamaindex/google@0.0.12
+  - @llamaindex/huggingface@0.0.41
+  - @llamaindex/mistral@0.0.10
+  - @llamaindex/mixedbread@0.0.10
+  - @llamaindex/ollama@0.0.45
+  - @llamaindex/openai@0.1.57
+  - @llamaindex/portkey-ai@0.0.38
+  - @llamaindex/replicate@0.0.38
+  - @llamaindex/astra@0.0.10
+  - @llamaindex/azure@0.1.5
+  - @llamaindex/chroma@0.0.10
+  - @llamaindex/firestore@1.0.3
+  - @llamaindex/milvus@0.1.5
+  - @llamaindex/mongodb@0.0.10
+  - @llamaindex/pinecone@0.0.10
+  - @llamaindex/postgres@0.0.38
+  - @llamaindex/qdrant@0.1.5
+  - @llamaindex/upstash@0.0.10
+  - @llamaindex/weaviate@0.0.10
+  - @llamaindex/vercel@0.0.16
+  - @llamaindex/readers@2.0.5
+  - @llamaindex/groq@0.0.56
+  - @llamaindex/vllm@0.0.27
+
+## 0.2.4
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+  - @llamaindex/cloud@3.0.5
+  - llamaindex@0.9.5
+  - @llamaindex/node-parser@1.0.4
+  - @llamaindex/anthropic@0.2.2
+  - @llamaindex/clip@0.0.40
+  - @llamaindex/cohere@0.0.9
+  - @llamaindex/deepinfra@0.0.40
+  - @llamaindex/google@0.0.11
+  - @llamaindex/huggingface@0.0.40
+  - @llamaindex/mistral@0.0.9
+  - @llamaindex/mixedbread@0.0.9
+  - @llamaindex/ollama@0.0.44
+  - @llamaindex/openai@0.1.56
+  - @llamaindex/portkey-ai@0.0.37
+  - @llamaindex/replicate@0.0.37
+  - @llamaindex/astra@0.0.9
+  - @llamaindex/azure@0.1.4
+  - @llamaindex/chroma@0.0.9
+  - @llamaindex/firestore@1.0.2
+  - @llamaindex/milvus@0.1.4
+  - @llamaindex/mongodb@0.0.9
+  - @llamaindex/pinecone@0.0.9
+  - @llamaindex/postgres@0.0.37
+  - @llamaindex/qdrant@0.1.4
+  - @llamaindex/upstash@0.0.9
+  - @llamaindex/weaviate@0.0.9
+  - @llamaindex/vercel@0.0.15
+  - @llamaindex/voyage-ai@1.0.1
+  - @llamaindex/readers@2.0.4
+  - @llamaindex/groq@0.0.55
+  - @llamaindex/vllm@0.0.26
+
+## 0.2.3
+
+### Patch Changes
+
+- Updated dependencies [cb256f2]
+- Updated dependencies [cb021e7]
+- Updated dependencies [0346391]
+  - @llamaindex/openai@0.1.55
+  - @llamaindex/core@0.5.3
+  - llamaindex@0.9.4
+  - @llamaindex/voyage-ai@1.0.0
+  - @llamaindex/clip@0.0.39
+  - @llamaindex/deepinfra@0.0.39
+  - @llamaindex/groq@0.0.54
+  - @llamaindex/huggingface@0.0.39
+  - @llamaindex/azure@0.1.3
+  - @llamaindex/milvus@0.1.3
+  - @llamaindex/qdrant@0.1.3
+  - @llamaindex/vllm@0.0.25
+  - @llamaindex/cloud@3.0.4
+  - @llamaindex/node-parser@1.0.3
+  - @llamaindex/anthropic@0.2.1
+  - @llamaindex/cohere@0.0.8
+  - @llamaindex/google@0.0.10
+  - @llamaindex/mistral@0.0.8
+  - @llamaindex/mixedbread@0.0.8
+  - @llamaindex/ollama@0.0.43
+  - @llamaindex/portkey-ai@0.0.36
+  - @llamaindex/replicate@0.0.36
+  - @llamaindex/astra@0.0.8
+  - @llamaindex/chroma@0.0.8
+  - @llamaindex/firestore@1.0.1
+  - @llamaindex/mongodb@0.0.8
+  - @llamaindex/pinecone@0.0.8
+  - @llamaindex/postgres@0.0.36
+  - @llamaindex/upstash@0.0.8
+  - @llamaindex/weaviate@0.0.8
+  - @llamaindex/vercel@0.0.14
+  - @llamaindex/readers@2.0.3
+
 ## 0.2.2
 
 ### Patch Changes

examples/agentworkflow/blog_writer.ts

@@ -0,0 +1,83 @@
+import { OpenAI } from "@llamaindex/openai";
+import fs from "fs";
+import {
+  AgentToolCall,
+  AgentToolCallResult,
+  AgentWorkflow,
+  FunctionAgent,
+  FunctionTool,
+} from "llamaindex";
+import os from "os";
+import { z } from "zod";
+
+import { WikipediaTool } from "../wiki";
+const llm = new OpenAI({
+  model: "gpt-4o-mini",
+});
+
+const saveFileTool = FunctionTool.from(
+  ({ content }: { content: string }) => {
+    const filePath = os.tmpdir() + "/report.md";
+    fs.writeFileSync(filePath, content);
+    return `File saved successfully at ${filePath}`;
+  },
+  {
+    name: "saveFile",
+    description:
+      "Save the written content into a file that can be downloaded by the user",
+    parameters: z.object({
+      content: z.string({
+        description: "The content to save into a file",
+      }),
+    }),
+  },
+);
+
+async function main() {
+  const reportAgent = new FunctionAgent({
+    name: "ReportAgent",
+    description:
+      "Responsible for crafting well-written blog posts based on research findings",
+    systemPrompt: `You are a professional writer. Your task is to create an engaging blog post using the research content provided. Once complete, save the post to a file using the saveFile tool.`,
+    tools: [saveFileTool],
+    llm,
+  });
+
+  const researchAgent = new FunctionAgent({
+    name: "ResearchAgent",
+    description:
+      "Responsible for gathering relevant information from the internet",
+    systemPrompt: `You are a research agent. Your role is to gather information from the internet using the provided tools and then transfer this information to the report agent for content creation.`,
+    tools: [new WikipediaTool()],
+    canHandoffTo: [reportAgent],
+    llm,
+  });
+
+  const workflow = new AgentWorkflow({
+    agents: [researchAgent, reportAgent],
+    rootAgent: researchAgent,
+  });
+
+  const context = workflow.run("Write a blog post about history of LLM");
+
+  let finalResult;
+  for await (const event of context) {
+    if (event instanceof AgentToolCall) {
+      console.log(
+        `[Agent ${event.displayName}] executing tool ${event.data.toolName} with parameters ${JSON.stringify(
+          event.data.toolKwargs,
+        )}`,
+      );
+    } else if (event instanceof AgentToolCallResult) {
+      console.log(
+        `[Agent ${event.displayName}] executed tool ${event.data.toolName} with result ${event.data.toolOutput.result}`,
+      );
+    }
+    finalResult = event;
+  }
+  console.log("Final result:", finalResult?.data);
+}
+
+main().catch((error) => {
+  console.error("Error:", error);
+});

examples/agentworkflow/multiple_agents.ts

@@ -0,0 +1,110 @@
+/**
+ * This example shows how to use AgentWorkflow with multiple agents
+ * 1. FetchWeatherAgent - Fetches the weather in a city
+ * 2. TemperatureConverterAgent - Converts the temperature from Fahrenheit to Celsius
+ */
+import { OpenAI } from "@llamaindex/openai";
+import { StopEvent } from "@llamaindex/workflow";
+import {
+  AgentInput,
+  AgentOutput,
+  AgentStream,
+  AgentToolCall,
+  AgentToolCallResult,
+  AgentWorkflow,
+  FunctionAgent,
+  FunctionTool,
+} from "llamaindex";
+import { z } from "zod";
+
+const llm = new OpenAI({
+  model: "gpt-4o-mini",
+});
+
+// Define tools for the agents
+const temperatureConverterTool = FunctionTool.from(
+  ({ temperature }: { temperature: number }) => {
+    return ((temperature - 32) * 5) / 9;
+  },
+  {
+    description: "Convert a temperature from Fahrenheit to Celsius",
+    name: "fahrenheitToCelsius",
+    parameters: z.object({
+      temperature: z.number({
+        description: "The temperature in Fahrenheit",
+      }),
+    }),
+  },
+);
+
+const temperatureFetcherTool = FunctionTool.from(
+  ({ city }: { city: string }) => {
+    const temperature = Math.floor(Math.random() * 58) + 32;
+    return `The current temperature in ${city} is ${temperature}°F`;
+  },
+  {
+    description: "Fetch the temperature (in Fahrenheit) for a city",
+    name: "fetchTemperature",
+    parameters: z.object({
+      city: z.string({
+        description: "The city to fetch the temperature for",
+      }),
+    }),
+  },
+);
+
+// Create agents
+async function multiWeatherAgent() {
+  const converterAgent = new FunctionAgent({
+    name: "TemperatureConverterAgent",
+    description:
+      "An agent that can convert temperatures from Fahrenheit to Celsius.",
+    tools: [temperatureConverterTool],
+    llm,
+  });
+
+  const weatherAgent = new FunctionAgent({
+    name: "FetchWeatherAgent",
+    description: "An agent that can get the weather in a city. ",
+    systemPrompt:
+      "If you can't answer the user question, hand off to other agents.",
+    tools: [temperatureFetcherTool],
+    llm,
+    // Define which next agents can be called next if this agent cannot complete the task
+    // Can be passed as agent name, e.g. "TemperatureConverterAgent"
+    canHandoffTo: [converterAgent],
+  });
+
+  // Create agent workflow with the agents
+  const workflow = new AgentWorkflow({
+    agents: [weatherAgent, converterAgent],
+    rootAgent: weatherAgent,
+    verbose: false,
+  });
+
+  // Ask the agent to get the weather in a city
+  const context = workflow.run(
+    "What is the weather in San Francisco in Celsius?",
+  );
+  // Stream the events
+  for await (const event of context) {
+    // These events might be useful for UI
+    if (
+      event instanceof AgentToolCall ||
+      event instanceof AgentToolCallResult ||
+      event instanceof AgentOutput ||
+      event instanceof AgentInput ||
+      event instanceof StopEvent
+    ) {
+      console.log(event);
+    } else if (event instanceof AgentStream) {
+      for (const chunk of event.data.delta) {
+        process.stdout.write(chunk);
+      }
+    }
+  }
+}
+
+multiWeatherAgent().catch((error) => {
+  console.error("Error:", error);
+});

examples/agentworkflow/single_agent.ts

@@ -0,0 +1,37 @@
+/**
+ * This example shows how to use AgentWorkflow as a single agent with tools
+ */
+import { OpenAI } from "@llamaindex/openai";
+import { AgentWorkflow } from "llamaindex";
+import { getWeatherTool } from "../agent/utils/tools";
+
+const llm = new OpenAI({
+  model: "gpt-4o",
+});
+
+async function singleWeatherAgent() {
+  const workflow = AgentWorkflow.fromTools({
+    tools: [getWeatherTool],
+    llm,
+    verbose: false,
+  });
+
+  const workflowContext = workflow.run(
+    "What's the weather like in San Francisco?",
+  );
+  const sfResult = await workflowContext;
+  // The weather in San Francisco, CA is currently sunny.
+  console.log(`${JSON.stringify(sfResult, null, 2)}`);
+
+  // Reuse the context from the previous run
+  const workflowContext2 = workflow.run("Compare it with California?", {
+    context: workflowContext.data,
+  });
+  const caResult = await workflowContext2;
+  // Both San Francisco and California are currently experiencing sunny weather.
+  console.log(`${JSON.stringify(caResult, null, 2)}`);
+}
+
+singleWeatherAgent().catch((error) => {
+  console.error("Error:", error);
+});

examples/agentworkflow/with-anthropic.ts

@@ -0,0 +1,114 @@
+import fs from "fs";
+import {
+  AgentToolCall,
+  AgentToolCallResult,
+  AgentWorkflow,
+  FunctionAgent,
+  FunctionTool,
+} from "llamaindex";
+import { z } from "zod";
+
+import { Anthropic } from "@llamaindex/anthropic";
+
+const llm = new Anthropic({
+  model: "claude-3-5-sonnet",
+});
+
+const weatherTool = FunctionTool.from(
+  (query: { location: string }) => {
+    return `The weather in ${query.location} is sunny`;
+  },
+  {
+    name: "weather",
+    description: "Get the weather",
+    parameters: z.object({
+      location: z.string({
+        description: "The location to get the weather for",
+      }),
+    }),
+  },
+);
+
+const inflationTool = FunctionTool.from(
+  (query: { location: string }) => {
+    return `The inflation in ${query.location} is 2%`;
+  },
+  {
+    name: "inflation",
+    description: "Get the inflation",
+    parameters: z.object({
+      location: z.string({
+        description: "The location to get the inflation for",
+      }),
+    }),
+  },
+);
+
+const saveFileTool = FunctionTool.from(
+  ({ content }: { content: string }) => {
+    const filePath = "./report.md";
+    fs.writeFileSync(filePath, content);
+    return `File saved successfully at ${filePath}`;
+  },
+  {
+    name: "saveFile",
+    description:
+      "Save the written content into a file that can be downloaded by the user",
+    parameters: z.object({
+      content: z.string({
+        description: "The content to save into a file",
+      }),
+    }),
+  },
+);
+
+async function main() {
+  const reportAgent = new FunctionAgent({
+    name: "ReportAgent",
+    description:
+      "Responsible for creating concise reports about weather and inflation data",
+    systemPrompt: `You are a professional writer. Your task is to create a clear and concise report summarizing the weather and inflation data provided. Once complete, save the report to a file using the saveFile tool.`,
+    tools: [saveFileTool],
+    llm,
+  });
+
+  const researchAgent = new FunctionAgent({
+    name: "ResearchAgent",
+    description:
+      "Responsible for gathering relevant information from the internet",
+    systemPrompt: `You are a research agent. Your role is to gather information about the inflation and weather in the location provided.`,
+    tools: [inflationTool, weatherTool],
+    canHandoffTo: [reportAgent],
+    llm,
+  });
+
+  const workflow = new AgentWorkflow({
+    agents: [researchAgent, reportAgent],
+    rootAgent: researchAgent,
+  });
+
+  const context = workflow.run(
+    "Write a report about New York weather and inflation",
+  );
+
+  let finalResult;
+  for await (const event of context) {
+    if (event instanceof AgentToolCall) {
+      console.log(
+        `[Agent ${event.displayName}] executing tool ${event.data.toolName} with parameters ${JSON.stringify(
+          event.data.toolKwargs,
+        )}`,
+      );
+    } else if (event instanceof AgentToolCallResult) {
+      console.log(
+        `[Agent ${event.displayName}] executed tool ${event.data.toolName} with result ${event.data.toolOutput.result}`,
+      );
+    }
+    finalResult = event;
+  }
+  console.log("Final result:", finalResult?.data);
+}
+
+main().catch((error) => {
+  console.error("Error:", error);
+});

examples/chat-engine/streaming.ts

@@ -0,0 +1,25 @@
+import { LLMAgent } from "llamaindex";
+import { stdin as input, stdout as output } from "node:process";
+import readline from "node:readline/promises";
+
+const agent = new LLMAgent({ tools: [] });
+
+(async () => {
+  const rl = readline.createInterface({ input, output });
+
+  while (true) {
+    const query = await rl.question("User: ");
+
+    const startTime = Date.now();
+    const stream = await agent.chat({ message: query, stream: true });
+    const timeToGetFirstChunk = Date.now() - startTime;
+    process.stdout.write(
+      `Time to get first chunk from LLMAgent: ${timeToGetFirstChunk}ms\n`,
+    );
+    process.stdout.write("Assistant with LLMAgent: ");
+    for await (const chunk of stream) {
+      process.stdout.write(chunk.response);
+    }
+    process.stdout.write("\n");
+  }
+})();

examples/deepseek.ts

@@ -0,0 +1,46 @@
+import { DeepSeekLLM } from "@llamaindex/deepseek";
+
+// process.env.DEEPSEEK_API_KEY is required
+const deepseek = new DeepSeekLLM({
+  apiKey: process.env.DEEPSEEK_API_KEY,
+  model: "deepseek-coder", // or "deepseek-chat"
+});
+
+(async () => {
+  // Example of non-streaming chat
+  const response = await deepseek.chat({
+    messages: [
+      {
+        role: "system",
+        content: "You are an AI assistant",
+      },
+      {
+        role: "user",
+        content: "Tell me about San Francisco",
+      },
+    ],
+    stream: false,
+  });
+  console.log("Response from DeepSeek AI:");
+  console.log(response);
+
+  // Example of streaming chat
+  const generator = await deepseek.chat({
+    messages: [
+      {
+        role: "system",
+        content: "You are an AI assistant",
+      },
+      {
+        role: "user",
+        content: "Write a short poem about San Francisco",
+      },
+    ],
+    stream: true,
+  });
+  console.log("\nStreaming response from DeepSeek AI...");
+  for await (const message of generator) {
+    process.stdout.write(message.delta);
+  }
+  console.log("\n");
+})();

examples/multimodal/jina.ts

@@ -1,10 +1,6 @@
+import { JinaAIEmbedding } from "@llamaindex/jinaai";
 import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
-import {
-  ImageDocument,
-  JinaAIEmbedding,
-  similarity,
-  SimilarityType,
-} from "llamaindex";
+import { ImageDocument, similarity, SimilarityType } from "llamaindex";
 import path from "path";
 
 async function main() {

examples/openai.ts

@@ -1,7 +1,7 @@
 import { OpenAI, OpenAIEmbedding } from "@llamaindex/openai";
 
 (async () => {
-  const llm = new OpenAI({ model: "gpt-4-1106-preview", temperature: 0.1 });
+  const llm = new OpenAI({ model: "gpt-4.5-preview", temperature: 0.1 });
 
   // complete api
   const response1 = await llm.complete({ prompt: "How are you?" });

examples/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/examples",
-  "version": "0.2.2",
+  "version": "0.2.8",
   "private": true,
   "scripts": {
     "lint": "eslint .",
@@ -11,38 +11,43 @@
     "@azure/cosmos": "^4.1.1",
     "@azure/identity": "^4.4.1",
     "@azure/search-documents": "^12.1.0",
-    "@llamaindex/anthropic": "^0.2.0",
-    "@llamaindex/astra": "^0.0.7",
-    "@llamaindex/azure": "^0.1.2",
-    "@llamaindex/chroma": "^0.0.7",
-    "@llamaindex/clip": "^0.0.38",
-    "@llamaindex/cloud": "^3.0.3",
-    "@llamaindex/cohere": "^0.0.7",
-    "@llamaindex/core": "^0.5.2",
-    "@llamaindex/deepinfra": "^0.0.38",
-    "@llamaindex/env": "^0.1.28",
-    "@llamaindex/firestore": "^1.0.0",
-    "@llamaindex/google": "^0.0.9",
-    "@llamaindex/groq": "^0.0.53",
-    "@llamaindex/huggingface": "^0.0.38",
-    "@llamaindex/milvus": "^0.1.2",
-    "@llamaindex/mistral": "^0.0.7",
-    "@llamaindex/mixedbread": "^0.0.7",
-    "@llamaindex/mongodb": "^0.0.7",
-    "@llamaindex/node-parser": "^1.0.2",
-    "@llamaindex/ollama": "^0.0.42",
-    "@llamaindex/openai": "^0.1.54",
-    "@llamaindex/pinecone": "^0.0.7",
-    "@llamaindex/portkey-ai": "^0.0.35",
-    "@llamaindex/postgres": "^0.0.35",
-    "@llamaindex/qdrant": "^0.1.2",
-    "@llamaindex/readers": "^2.0.2",
-    "@llamaindex/replicate": "^0.0.35",
-    "@llamaindex/upstash": "^0.0.7",
-    "@llamaindex/vercel": "^0.0.13",
-    "@llamaindex/vllm": "^0.0.24",
-    "@llamaindex/weaviate": "^0.0.7",
-    "@llamaindex/workflow": "^0.0.11",
+    "@llamaindex/anthropic": "^0.2.5",
+    "@llamaindex/astra": "^0.0.12",
+    "@llamaindex/azure": "^0.1.7",
+    "@llamaindex/chroma": "^0.0.12",
+    "@llamaindex/clip": "^0.0.43",
+    "@llamaindex/cloud": "^3.0.8",
+    "@llamaindex/cohere": "^0.0.12",
+    "@llamaindex/core": "^0.5.7",
+    "@llamaindex/deepinfra": "^0.0.43",
+    "@llamaindex/env": "^0.1.29",
+    "@llamaindex/firestore": "^1.0.5",
+    "@llamaindex/google": "^0.1.0",
+    "@llamaindex/groq": "^0.0.58",
+    "@llamaindex/huggingface": "^0.0.43",
+    "@llamaindex/milvus": "^0.1.7",
+    "@llamaindex/mistral": "^0.0.12",
+    "@llamaindex/mixedbread": "^0.0.12",
+    "@llamaindex/mongodb": "^0.0.12",
+    "@llamaindex/node-parser": "^1.0.7",
+    "@llamaindex/ollama": "^0.0.47",
+    "@llamaindex/openai": "^0.1.59",
+    "@llamaindex/pinecone": "^0.0.12",
+    "@llamaindex/portkey-ai": "^0.0.40",
+    "@llamaindex/postgres": "^0.0.40",
+    "@llamaindex/qdrant": "^0.1.7",
+    "@llamaindex/readers": "^2.0.7",
+    "@llamaindex/replicate": "^0.0.40",
+    "@llamaindex/upstash": "^0.0.12",
+    "@llamaindex/vercel": "^0.0.18",
+    "@llamaindex/vllm": "^0.0.29",
+    "@llamaindex/voyage-ai": "^1.0.4",
+    "@llamaindex/weaviate": "^0.0.12",
+    "@llamaindex/workflow": "^0.0.15",
+    "@llamaindex/deepseek": "^0.0.3",
+    "@llamaindex/fireworks": "^0.0.3",
+    "@llamaindex/together": "^0.0.3",
+    "@llamaindex/jinaai": "^0.0.3",
     "@notionhq/client": "^2.2.15",
     "@pinecone-database/pinecone": "^4.0.0",
     "@vercel/postgres": "^0.10.0",
@@ -51,11 +56,11 @@
     "commander": "^12.1.0",
     "dotenv": "^16.4.5",
     "js-tiktoken": "^1.0.14",
-    "llamaindex": "^0.9.3",
+    "llamaindex": "^0.9.9",
     "mongodb": "6.7.0",
-    "pathe": "^1.1.2",
     "postgres": "^3.4.4",
-    "wikipedia": "^2.1.2"
+    "wikipedia": "^2.1.2",
+    "zod": "^3.23.8"
   },
   "devDependencies": {
     "@types/node": "^22.9.0",

examples/qdrantdb/with-gemini.ts

@@ -0,0 +1,33 @@
+import {
+  GEMINI_EMBEDDING_MODEL,
+  GeminiEmbedding,
+  GeminiSession,
+} from "@llamaindex/google";
+import { QdrantVectorStore } from "@llamaindex/qdrant";
+import {
+  Document,
+  storageContextFromDefaults,
+  VectorStoreIndex,
+} from "llamaindex";
+
+const embedding = new GeminiEmbedding({
+  model: GEMINI_EMBEDDING_MODEL.EMBEDDING_001,
+  session: new GeminiSession({
+    apiKey: process.env.GEMINI_API_KEY,
+  }),
+});
+
+async function main() {
+  const docs = [new Document({ text: "Lorem ipsum dolor sit amet" })];
+  const vectorStore = new QdrantVectorStore({
+    url: process.env.QDRANT_URL,
+    apiKey: process.env.QDRANT_API_KEY,
+    embeddingModel: embedding,
+    collectionName: "gemini_test",
+  });
+  const storageContext = await storageContextFromDefaults({ vectorStore });
+  await VectorStoreIndex.fromDocuments(docs, { storageContext });
+  console.log("Inizialized vector store successfully");
+}
+
+void main().catch((err) => console.error(err));

examples/qdrantdb/with-jina.ts

@@ -0,0 +1,27 @@
+import { JinaAIEmbedding } from "@llamaindex/jinaai";
+import { QdrantVectorStore } from "@llamaindex/qdrant";
+import {
+  Document,
+  storageContextFromDefaults,
+  VectorStoreIndex,
+} from "llamaindex";
+
+const embedding = new JinaAIEmbedding({
+  apiKey: process.env.JINAAI_API_KEY,
+  model: "jina-embeddings-v3",
+});
+
+async function main() {
+  const docs = [new Document({ text: "Lorem ipsum dolor sit amet" })];
+  const vectorStore = new QdrantVectorStore({
+    url: process.env.QDRANT_URL,
+    apiKey: process.env.QDRANT_API_KEY,
+    embeddingModel: embedding,
+    collectionName: "jina_test",
+  });
+  const storageContext = await storageContextFromDefaults({ vectorStore });
+  await VectorStoreIndex.fromDocuments(docs, { storageContext });
+  console.log("Inizialized vector store successfully");
+}
+
+void main().catch((err) => console.error(err));

examples/readers/src/pdf_fw.ts

@@ -1,5 +1,6 @@
+import { FireworksEmbedding, FireworksLLM } from "@llamaindex/fireworks";
 import { PDFReader } from "@llamaindex/readers/pdf";
-import { FireworksEmbedding, FireworksLLM, VectorStoreIndex } from "llamaindex";
+import { VectorStoreIndex } from "llamaindex";
 
 import { Settings } from "llamaindex";
 

examples/together-ai.ts

@@ -1,4 +1,4 @@
-import { TogetherEmbedding, TogetherLLM } from "llamaindex";
+import { TogetherEmbedding, TogetherLLM } from "@llamaindex/together";
 
 // process.env.TOGETHER_API_KEY is required
 const together = new TogetherLLM({

examples/together-ai/vector-index.ts

@@ -1,12 +1,7 @@
 import fs from "node:fs/promises";
 
-import {
-  Document,
-  Settings,
-  TogetherEmbedding,
-  TogetherLLM,
-  VectorStoreIndex,
-} from "llamaindex";
+import { TogetherEmbedding, TogetherLLM } from "@llamaindex/together";
+import { Document, Settings, VectorStoreIndex } from "llamaindex";
 
 // Update llm to use TogetherAI
 Settings.llm = new TogetherLLM({

examples/toolsStream.ts

@@ -15,7 +15,7 @@ async function main() {
     tools: [
       {
         metadata: {
-          name: "wikipedia_tool",
+          name: "wikipedia_search",
           description: "A tool that uses a query engine to search Wikipedia.",
           parameters: {
             type: "object",

examples/voyage-ai/embedding.ts

@@ -0,0 +1,17 @@
+import { VoyageAIEmbedding } from "@llamaindex/voyage-ai";
+
+async function main() {
+  // API token can be provided as an environment variable too
+  // using VOYAGE_API_TOKEN variable
+  const apiKey = process.env.VOYAGE_API_TOKEN ?? "YOUR_API_TOKEN";
+  const model = "voyage-3-lite";
+  const embedModel = new VoyageAIEmbedding({
+    model,
+    apiKey,
+  });
+  const texts = ["hello", "world"];
+  const embeddings = await embedModel.getTextEmbeddingsBatch(texts);
+  console.log(`\nWe have ${embeddings.length} embeddings`);
+}
+
+main().catch(console.error);

examples/wiki.ts

@@ -14,7 +14,7 @@ type WikipediaToolParams = {
 };
 
 const DEFAULT_META_DATA: ToolMetadata<JSONSchemaType<WikipediaParameter>> = {
-  name: "wikipedia_tool",
+  name: "wikipedia_search",
   description: "A tool that uses a query engine to search Wikipedia.",
   parameters: {
     type: "object",

packages/autotool/CHANGELOG.md

@@ -1,5 +1,44 @@
 # @llamaindex/autotool
 
+## 6.0.9
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 6.0.8
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 6.0.7
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+
+## 6.0.6
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 6.0.5
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 6.0.4
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 6.0.3
 
 ### Patch Changes

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

@@ -1,5 +1,50 @@
 # @llamaindex/autotool-01-node-example
 
+## 0.0.90
+
+### Patch Changes
+
+- llamaindex@0.9.9
+- @llamaindex/autotool@6.0.9
+
+## 0.0.89
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+  - @llamaindex/autotool@6.0.8
+
+## 0.0.88
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+  - @llamaindex/autotool@6.0.7
+
+## 0.0.87
+
+### Patch Changes
+
+- llamaindex@0.9.6
+- @llamaindex/autotool@6.0.6
+
+## 0.0.86
+
+### Patch Changes
+
+- llamaindex@0.9.5
+- @llamaindex/autotool@6.0.5
+
+## 0.0.85
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+  - @llamaindex/autotool@6.0.4
+
 ## 0.0.84
 
 ### Patch Changes

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

@@ -13,5 +13,5 @@
   "scripts": {
     "start": "node --import tsx --import @llamaindex/autotool/node ./src/index.ts"
   },
-  "version": "0.0.84"
+  "version": "0.0.90"
 }

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

@@ -1,5 +1,50 @@
 # @llamaindex/autotool-02-next-example
 
+## 0.1.134
+
+### Patch Changes
+
+- llamaindex@0.9.9
+- @llamaindex/autotool@6.0.9
+
+## 0.1.133
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+  - @llamaindex/autotool@6.0.8
+
+## 0.1.132
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+  - @llamaindex/autotool@6.0.7
+
+## 0.1.131
+
+### Patch Changes
+
+- llamaindex@0.9.6
+- @llamaindex/autotool@6.0.6
+
+## 0.1.130
+
+### Patch Changes
+
+- llamaindex@0.9.5
+- @llamaindex/autotool@6.0.5
+
+## 0.1.129
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+  - @llamaindex/autotool@6.0.4
+
 ## 0.1.128
 
 ### Patch Changes

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

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool-02-next-example",
   "private": true,
-  "version": "0.1.128",
+  "version": "0.1.134",
   "scripts": {
     "dev": "next dev",
     "build": "next build",

packages/autotool/package.json

@@ -6,7 +6,7 @@
     "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
     "directory": "packages/autotool"
   },
-  "version": "6.0.3",
+  "version": "6.0.9",
   "description": "auto transpile your JS function to LLM Agent compatible",
   "files": [
     "dist",

packages/cloud/CHANGELOG.md

@@ -1,5 +1,41 @@
 # @llamaindex/cloud
 
+## 3.0.8
+
+### Patch Changes
+
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+
+## 3.0.7
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+
+## 3.0.6
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+
+## 3.0.5
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+
+## 3.0.4
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - @llamaindex/core@0.5.3
+
 ## 3.0.3
 
 ### Patch Changes

packages/cloud/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloud",
-  "version": "3.0.3",
+  "version": "3.0.8",
   "type": "module",
   "license": "MIT",
   "scripts": {

packages/community/CHANGELOG.md

@@ -1,5 +1,43 @@
 # @llamaindex/community
 
+## 0.0.89
+
+### Patch Changes
+
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+
+## 0.0.88
+
+### Patch Changes
+
+- e28c29d: Added Llama 3.3 70B Instruct support
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+
+## 0.0.87
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+
+## 0.0.86
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+
+## 0.0.85
+
+### Patch Changes
+
+- 1914b52: Added Claude 3.7 Sonnet support
+- Updated dependencies [cb021e7]
+  - @llamaindex/core@0.5.3
+
 ## 0.0.84
 
 ### Patch Changes

packages/community/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/community",
   "description": "Community package for LlamaIndexTS",
-  "version": "0.0.84",
+  "version": "0.0.89",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

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

@@ -6,7 +6,19 @@ export type ToolChoice =
   | { type: "auto" }
   | { type: "tool"; name: string };
 
-export type AnthropicAdditionalChatOptions = { toolChoice: ToolChoice };
+export interface ThinkingConfigDisabled {
+  type: "disabled";
+}
+
+export interface ThinkingConfigEnabled {
+  budget_tokens: number;
+  type: "enabled";
+}
+
+export type AnthropicAdditionalChatOptions = {
+  toolChoice: ToolChoice;
+  thinking?: ThinkingConfigDisabled | ThinkingConfigEnabled;
+};
 
 type Usage = {
   input_tokens: number;

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

@@ -69,6 +69,7 @@ export const BEDROCK_MODELS = {
   ANTHROPIC_CLAUDE_3_5_SONNET: "anthropic.claude-3-5-sonnet-20240620-v1:0",
   ANTHROPIC_CLAUDE_3_5_SONNET_V2: "anthropic.claude-3-5-sonnet-20241022-v2:0",
   ANTHROPIC_CLAUDE_3_5_HAIKU: "anthropic.claude-3-5-haiku-20241022-v1:0",
+  ANTHROPIC_CLAUDE_3_7_SONNET: "anthropic.claude-3-7-sonnet-20250219-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",
@@ -80,6 +81,7 @@ export const BEDROCK_MODELS = {
   META_LLAMA3_2_3B_INSTRUCT: "meta.llama3-2-3b-instruct-v1:0",
   META_LLAMA3_2_11B_INSTRUCT: "meta.llama3-2-11b-instruct-v1:0",
   META_LLAMA3_2_90B_INSTRUCT: "meta.llama3-2-90b-instruct-v1:0",
+  META_LLAMA3_3_70B_INSTRUCT: "meta.llama3-3-70b-instruct-v1:0",
   MISTRAL_7B_INSTRUCT: "mistral.mistral-7b-instruct-v0:2",
   MISTRAL_MIXTRAL_7B_INSTRUCT: "mistral.mixtral-8x7b-instruct-v0:1",
   MISTRAL_MIXTRAL_LARGE_2402: "mistral.mistral-large-2402-v1:0",
@@ -100,10 +102,13 @@ export const INFERENCE_BEDROCK_MODELS = {
     "us.anthropic.claude-3-5-sonnet-20240620-v1:0",
   US_ANTHROPIC_CLAUDE_3_5_SONNET_V2:
     "us.anthropic.claude-3-5-sonnet-20241022-v2:0",
+  US_ANTHROPIC_CLAUDE_3_7_SONNET:
+    "us.anthropic.claude-3-7-sonnet-20250219-v1:0",
   US_META_LLAMA_3_2_1B_INSTRUCT: "us.meta.llama3-2-1b-instruct-v1:0",
   US_META_LLAMA_3_2_3B_INSTRUCT: "us.meta.llama3-2-3b-instruct-v1:0",
   US_META_LLAMA_3_2_11B_INSTRUCT: "us.meta.llama3-2-11b-instruct-v1:0",
   US_META_LLAMA_3_2_90B_INSTRUCT: "us.meta.llama3-2-90b-instruct-v1:0",
+  US_META_LLAMA_3_3_70B_INSTRUCT: "us.meta.llama3-3-70b-instruct-v1:0",
   US_AMAZON_NOVA_PRO_1: "us.amazon.nova-pro-v1:0",
   US_AMAZON_NOVA_LITE_1: "us.amazon.nova-lite-v1:0",
   US_AMAZON_NOVA_MICRO_1: "us.amazon.nova-micro-v1:0",
@@ -113,6 +118,8 @@ export const INFERENCE_BEDROCK_MODELS = {
   EU_ANTHROPIC_CLAUDE_3_SONNET: "eu.anthropic.claude-3-sonnet-20240229-v1:0",
   EU_ANTHROPIC_CLAUDE_3_5_SONNET:
     "eu.anthropic.claude-3-5-sonnet-20240620-v1:0",
+  EU_ANTHROPIC_CLAUDE_3_7_SONNET:
+    "eu.anthropic.claude-3-7-sonnet-20250219-v1:0",
   EU_META_LLAMA_3_2_1B_INSTRUCT: "eu.meta.llama3-2-1b-instruct-v1:0",
   EU_META_LLAMA_3_2_3B_INSTRUCT: "eu.meta.llama3-2-3b-instruct-v1:0",
 };
@@ -132,6 +139,8 @@ export const INFERENCE_TO_BEDROCK_MAP: Record<
     BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_SONNET,
   [INFERENCE_BEDROCK_MODELS.US_ANTHROPIC_CLAUDE_3_5_SONNET]:
     BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
+  [INFERENCE_BEDROCK_MODELS.US_ANTHROPIC_CLAUDE_3_7_SONNET]:
+    BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_7_SONNET,
   [INFERENCE_BEDROCK_MODELS.US_ANTHROPIC_CLAUDE_3_5_SONNET_V2]:
     BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET_V2,
   [INFERENCE_BEDROCK_MODELS.US_ANTHROPIC_CLAUDE_3_5_HAIKU]:
@@ -144,6 +153,8 @@ export const INFERENCE_TO_BEDROCK_MAP: Record<
     BEDROCK_MODELS.META_LLAMA3_2_11B_INSTRUCT,
   [INFERENCE_BEDROCK_MODELS.US_META_LLAMA_3_2_90B_INSTRUCT]:
     BEDROCK_MODELS.META_LLAMA3_2_90B_INSTRUCT,
+  [INFERENCE_BEDROCK_MODELS.US_META_LLAMA_3_3_70B_INSTRUCT]:
+    BEDROCK_MODELS.META_LLAMA3_3_70B_INSTRUCT,
 
   [INFERENCE_BEDROCK_MODELS.US_AMAZON_NOVA_PRO_1]:
     BEDROCK_MODELS.AMAZON_NOVA_PRO_1,
@@ -158,6 +169,8 @@ export const INFERENCE_TO_BEDROCK_MAP: Record<
     BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_SONNET,
   [INFERENCE_BEDROCK_MODELS.EU_ANTHROPIC_CLAUDE_3_5_SONNET]:
     BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
+  [INFERENCE_BEDROCK_MODELS.EU_ANTHROPIC_CLAUDE_3_7_SONNET]:
+    BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_7_SONNET,
   [INFERENCE_BEDROCK_MODELS.EU_META_LLAMA_3_2_1B_INSTRUCT]:
     BEDROCK_MODELS.META_LLAMA3_2_1B_INSTRUCT,
   [INFERENCE_BEDROCK_MODELS.EU_META_LLAMA_3_2_3B_INSTRUCT]:
@@ -191,6 +204,7 @@ const CHAT_ONLY_MODELS = {
   [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET]: 200000,
   [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET_V2]: 200000,
   [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_HAIKU]: 200000,
+  [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_7_SONNET]: 200000,
   [BEDROCK_MODELS.META_LLAMA2_13B_CHAT]: 2048,
   [BEDROCK_MODELS.META_LLAMA2_70B_CHAT]: 4096,
   [BEDROCK_MODELS.META_LLAMA3_8B_INSTRUCT]: 8192,
@@ -202,6 +216,7 @@ const CHAT_ONLY_MODELS = {
   [BEDROCK_MODELS.META_LLAMA3_2_3B_INSTRUCT]: 131000,
   [BEDROCK_MODELS.META_LLAMA3_2_11B_INSTRUCT]: 128000,
   [BEDROCK_MODELS.META_LLAMA3_2_90B_INSTRUCT]: 128000,
+  [BEDROCK_MODELS.META_LLAMA3_3_70B_INSTRUCT]: 128000,
   [BEDROCK_MODELS.MISTRAL_7B_INSTRUCT]: 32000,
   [BEDROCK_MODELS.MISTRAL_MIXTRAL_7B_INSTRUCT]: 32000,
   [BEDROCK_MODELS.MISTRAL_MIXTRAL_LARGE_2402]: 32000,
@@ -230,6 +245,7 @@ export const STREAMING_MODELS = new Set([
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET_V2,
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_HAIKU,
+  BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_7_SONNET,
   BEDROCK_MODELS.META_LLAMA2_13B_CHAT,
   BEDROCK_MODELS.META_LLAMA2_70B_CHAT,
   BEDROCK_MODELS.META_LLAMA3_8B_INSTRUCT,
@@ -241,6 +257,7 @@ export const STREAMING_MODELS = new Set([
   BEDROCK_MODELS.META_LLAMA3_2_3B_INSTRUCT,
   BEDROCK_MODELS.META_LLAMA3_2_11B_INSTRUCT,
   BEDROCK_MODELS.META_LLAMA3_2_90B_INSTRUCT,
+  BEDROCK_MODELS.META_LLAMA3_3_70B_INSTRUCT,
   BEDROCK_MODELS.MISTRAL_7B_INSTRUCT,
   BEDROCK_MODELS.MISTRAL_MIXTRAL_7B_INSTRUCT,
   BEDROCK_MODELS.MISTRAL_MIXTRAL_LARGE_2402,
@@ -256,11 +273,13 @@ export const TOOL_CALL_MODELS: BEDROCK_MODELS[] = [
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET_V2,
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_HAIKU,
+  BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_7_SONNET,
   BEDROCK_MODELS.META_LLAMA3_1_405B_INSTRUCT,
   BEDROCK_MODELS.META_LLAMA3_2_1B_INSTRUCT,
   BEDROCK_MODELS.META_LLAMA3_2_3B_INSTRUCT,
   BEDROCK_MODELS.META_LLAMA3_2_11B_INSTRUCT,
   BEDROCK_MODELS.META_LLAMA3_2_90B_INSTRUCT,
+  BEDROCK_MODELS.META_LLAMA3_3_70B_INSTRUCT,
   BEDROCK_MODELS.AMAZON_NOVA_PRO_1,
   BEDROCK_MODELS.AMAZON_NOVA_LITE_1,
   BEDROCK_MODELS.AMAZON_NOVA_MICRO_1,
@@ -294,6 +313,7 @@ export const BEDROCK_MODEL_MAX_TOKENS: Partial<Record<BEDROCK_MODELS, number>> =
     [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET]: 4096,
     [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET_V2]: 8192,
     [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_HAIKU]: 8192,
+    [BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_7_SONNET]: 8192,
     [BEDROCK_MODELS.META_LLAMA2_13B_CHAT]: 2048,
     [BEDROCK_MODELS.META_LLAMA2_70B_CHAT]: 2048,
     [BEDROCK_MODELS.META_LLAMA3_8B_INSTRUCT]: 2048,
@@ -305,6 +325,7 @@ export const BEDROCK_MODEL_MAX_TOKENS: Partial<Record<BEDROCK_MODELS, number>> =
     [BEDROCK_MODELS.META_LLAMA3_2_3B_INSTRUCT]: 2048,
     [BEDROCK_MODELS.META_LLAMA3_2_11B_INSTRUCT]: 2048,
     [BEDROCK_MODELS.META_LLAMA3_2_90B_INSTRUCT]: 2048,
+    [BEDROCK_MODELS.META_LLAMA3_3_70B_INSTRUCT]: 2048,
   };
 
 const DEFAULT_BEDROCK_PARAMS = {

packages/core/CHANGELOG.md

@@ -1,5 +1,37 @@
 # @llamaindex/core
 
+## 0.5.7
+
+### Patch Changes
+
+- 4bac71d: Support binding additional argument to function tool
+
+## 0.5.6
+
+### Patch Changes
+
+- beb922b: Fix edge runtime builds by adding missing packages to env package. Make gpt-tokenizer optional for llamaindex to reduce package size.
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+
+## 0.5.5
+
+### Patch Changes
+
+- 5668970: feat: Support AgentWorkflow
+
+## 0.5.4
+
+### Patch Changes
+
+- ad3c7f1: fix: streaming issues with LLMAgent
+
+## 0.5.3
+
+### Patch Changes
+
+- cb021e7: feat(node-parser): support async function
+
 ## 0.5.2
 
 ### Patch Changes

packages/core/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/core",
   "type": "module",
-  "version": "0.5.2",
+  "version": "0.5.7",
   "description": "LlamaIndex Core Module",
   "exports": {
     "./agent": {

packages/core/src/agent/base.ts

@@ -193,7 +193,7 @@ export abstract class AgentWorker<
     return new ReadableStream<
       TaskStepOutput<AI, Store, AdditionalMessageOptions, AdditionalChatOptions>
     >({
-      start: async (controller) => {
+      pull: async (controller) => {
         for await (const stepOutput of taskOutputStream) {
           this.#taskSet.add(stepOutput.taskStep);
           if (stepOutput.isLast) {
@@ -209,23 +209,31 @@ export abstract class AgentWorker<
             }
             const { output, taskStep } = stepOutput;
             if (output instanceof ReadableStream) {
-              const [pipStream, finalStream] = output.tee();
-              stepOutput.output = finalStream;
-              const reader = pipStream.getReader();
-              const { value } = await reader.read();
-              reader.releaseLock();
-              let content: string = value!.delta;
-              for await (const chunk of pipStream) {
-                content += chunk.delta;
-              }
-              taskStep.context.store.messages = [
-                ...taskStep.context.store.messages,
-                {
-                  role: "assistant",
-                  content,
-                  options: value!.options,
-                },
-              ];
+              let content = "";
+              let options: AdditionalMessageOptions | undefined = undefined;
+              const transformedStream = output.pipeThrough(
+                new TransformStream({
+                  transform(chunk, controller) {
+                    content += chunk.delta;
+                    if (!options && chunk.options) {
+                      options = chunk.options;
+                    }
+                    controller.enqueue(chunk); // Pass the chunk through unchanged
+                  },
+                  // When stream finishes, store the accumulated message in context
+                  flush() {
+                    taskStep.context.store.messages = [
+                      ...taskStep.context.store.messages,
+                      {
+                        role: "assistant",
+                        content,
+                        options,
+                      },
+                    ];
+                  },
+                }),
+              );
+              stepOutput.output = transformedStream;
             }
             controller.enqueue(stepOutput);
             controller.close();

packages/core/src/llms/type.ts

@@ -215,13 +215,19 @@ export type ToolMetadata<
    * @link https://json-schema.org/understanding-json-schema
    */
   parameters?: Parameters;
+  /**
+   * Whether the tool requires workflow context to be passed in.
+   */
+  requireContext?: boolean;
 };
 
 /**
  * Simple Tool interface. Likely to change.
  */
-// eslint-disable-next-line @typescript-eslint/no-explicit-any
-export interface BaseTool<Input = any> {
+export interface BaseTool<
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  Input = any,
+> {
   /**
    * This could be undefined if the implementation is not provided,
    *  which might be the case when communicating with a llm.

packages/core/src/node-parser/base.ts

@@ -7,21 +7,27 @@ import {
   TextNode,
   TransformComponent,
 } from "../schema";
+import { isPromise } from "../utils";
 
-export abstract class NodeParser extends TransformComponent<BaseNode[]> {
+export abstract class NodeParser<
+  Result extends TextNode[] | Promise<TextNode[]> =
+    | TextNode[]
+    | Promise<TextNode[]>,
+> extends TransformComponent<Result> {
   includeMetadata: boolean = true;
   includePrevNextRel: boolean = true;
 
   constructor() {
-    super((nodes: BaseNode[]): BaseNode[] => {
+    super((nodes: BaseNode[]): Result => {
+      // alex: should we fix `as` type?
       return this.getNodesFromDocuments(nodes as TextNode[]);
     });
   }
 
   protected postProcessParsedNodes(
-    nodes: TextNode[],
+    nodes: Awaited<Result>,
     parentDocMap: Map<string, TextNode>,
-  ): TextNode[] {
+  ): Awaited<Result> {
     nodes.forEach((node, i) => {
       const parentDoc = parentDocMap.get(node.sourceNode?.nodeId || "");
 
@@ -73,9 +79,9 @@ export abstract class NodeParser extends TransformComponent<BaseNode[]> {
   protected abstract parseNodes(
     documents: TextNode[],
     showProgress?: boolean,
-  ): TextNode[];
+  ): Result;
 
-  public getNodesFromDocuments(documents: TextNode[]): TextNode[] {
+  public getNodesFromDocuments(documents: TextNode[]): Result {
     const docsId: Map<string, TextNode> = new Map(
       documents.map((doc) => [doc.id_, doc]),
     );
@@ -85,20 +91,36 @@ export abstract class NodeParser extends TransformComponent<BaseNode[]> {
       documents,
     });
 
-    const nodes = this.postProcessParsedNodes(
-      this.parseNodes(documents),
-      docsId,
-    );
+    const parsedNodes = this.parseNodes(documents);
+    if (isPromise(parsedNodes)) {
+      return parsedNodes.then((parsedNodes) => {
+        const nodes = this.postProcessParsedNodes(
+          parsedNodes as Awaited<Result>,
+          docsId,
+        );
 
-    callbackManager.dispatchEvent("node-parsing-end", {
-      nodes,
-    });
+        callbackManager.dispatchEvent("node-parsing-end", {
+          nodes,
+        });
 
-    return nodes;
+        return nodes;
+      }) as Result;
+    } else {
+      const nodes = this.postProcessParsedNodes(
+        parsedNodes as Awaited<Result>,
+        docsId,
+      );
+
+      callbackManager.dispatchEvent("node-parsing-end", {
+        nodes,
+      });
+
+      return nodes;
+    }
   }
 }
 
-export abstract class TextSplitter extends NodeParser {
+export abstract class TextSplitter extends NodeParser<TextNode[]> {
   abstract splitText(text: string): string[];
 
   public splitTexts(texts: string[]): string[] {

packages/core/src/node-parser/markdown.ts

@@ -6,7 +6,7 @@ import {
 } from "../schema";
 import { NodeParser } from "./base";
 
-export class MarkdownNodeParser extends NodeParser {
+export class MarkdownNodeParser extends NodeParser<TextNode[]> {
   override parseNodes(nodes: TextNode[], showProgress?: boolean): TextNode[] {
     return nodes.reduce<TextNode[]>((allNodes, node) => {
       const markdownNodes = this.getNodesFromNode(node);

packages/core/src/node-parser/sentence-window.ts

@@ -9,7 +9,7 @@ import {
 import { NodeParser } from "./base";
 import { splitBySentenceTokenizer, type TextSplitterFn } from "./utils";
 
-export class SentenceWindowNodeParser extends NodeParser {
+export class SentenceWindowNodeParser extends NodeParser<TextNode[]> {
   static DEFAULT_WINDOW_SIZE = 3;
   static DEFAULT_WINDOW_METADATA_KEY = "window";
   static DEFAULT_ORIGINAL_TEXT_METADATA_KEY = "originalText";

packages/core/src/tools/function-tool.ts

@@ -4,40 +4,66 @@ import { zodToJsonSchema } from "zod-to-json-schema";
 import type { JSONValue } from "../global";
 import type { BaseTool, ToolMetadata } from "../llms";
 
-export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
-  implements BaseTool<T>
+export class FunctionTool<
+  T,
+  R extends JSONValue | Promise<JSONValue>,
+  AdditionalToolArgument extends object = object,
+> implements BaseTool<T>
 {
-  #fn: (input: T) => R;
+  #fn: (input: T, additionalArg?: AdditionalToolArgument) => R;
+  #additionalArg: AdditionalToolArgument | undefined;
   readonly #metadata: ToolMetadata<JSONSchemaType<T>>;
   readonly #zodType: z.ZodType<T> | null = null;
   constructor(
-    fn: (input: T) => R,
+    fn: (input: T, additionalArg?: AdditionalToolArgument) => R,
     metadata: ToolMetadata<JSONSchemaType<T>>,
     zodType?: z.ZodType<T>,
+    additionalArg?: AdditionalToolArgument,
   ) {
     this.#fn = fn;
     this.#metadata = metadata;
     if (zodType) {
       this.#zodType = zodType;
     }
+    this.#additionalArg = additionalArg;
   }
 
-  static from<T>(
-    fn: (input: T) => JSONValue | Promise<JSONValue>,
+  static from<T, AdditionalToolArgument extends object = object>(
+    fn: (
+      input: T,
+      additionalArg?: AdditionalToolArgument,
+    ) => JSONValue | Promise<JSONValue>,
     schema: ToolMetadata<JSONSchemaType<T>>,
-  ): FunctionTool<T, JSONValue | Promise<JSONValue>>;
-  static from<R extends z.ZodType>(
-    fn: (input: z.infer<R>) => JSONValue | Promise<JSONValue>,
+  ): FunctionTool<T, JSONValue | Promise<JSONValue>, AdditionalToolArgument>;
+  static from<
+    R extends z.ZodType,
+    AdditionalToolArgument extends object = object,
+  >(
+    fn: (
+      input: z.infer<R>,
+      additionalArg?: AdditionalToolArgument,
+    ) => JSONValue | Promise<JSONValue>,
     schema: Omit<ToolMetadata, "parameters"> & {
       parameters: R;
     },
-  ): FunctionTool<z.infer<R>, JSONValue | Promise<JSONValue>>;
-  static from<T, R extends z.ZodType<T>>(
-    fn: (input: T) => JSONValue | Promise<JSONValue>,
+  ): FunctionTool<
+    z.infer<R>,
+    JSONValue | Promise<JSONValue>,
+    AdditionalToolArgument
+  >;
+  static from<
+    T,
+    R extends z.ZodType<T>,
+    AdditionalToolArgument extends object = object,
+  >(
+    fn: (
+      input: T,
+      additionalArg?: AdditionalToolArgument,
+    ) => JSONValue | Promise<JSONValue>,
     schema: Omit<ToolMetadata, "parameters"> & {
       parameters: R;
     },
-  ): FunctionTool<T, JSONValue>;
+  ): FunctionTool<T, JSONValue, AdditionalToolArgument>;
   // eslint-disable-next-line @typescript-eslint/no-explicit-any
   static from(fn: any, schema: any): any {
     if (schema.parameters instanceof z.ZodSchema) {
@@ -58,15 +84,41 @@ export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
     return this.#metadata as BaseTool<T>["metadata"];
   }
 
+  bind = (additionalArg: AdditionalToolArgument) => {
+    return new FunctionTool(
+      this.#fn,
+      this.#metadata,
+      this.#zodType ?? undefined,
+      additionalArg,
+    );
+  };
+
   call = (input: T) => {
+    if (this.#metadata.requireContext) {
+      const inputWithContext = input as Record<string, unknown>;
+      if (!inputWithContext.context) {
+        throw new Error(
+          "Tool call requires context, but context parameter is missing",
+        );
+      }
+    }
     if (this.#zodType) {
       const result = this.#zodType.safeParse(input);
       if (result.success) {
-        return this.#fn.call(null, result.data);
+        if (this.#metadata.requireContext) {
+          const { context } = input as Record<string, unknown>;
+          return this.#fn.call(
+            null,
+            { context, ...result.data },
+            this.#additionalArg,
+          );
+        } else {
+          return this.#fn.call(null, result.data, this.#additionalArg);
+        }
       } else {
         console.warn(result.error.errors);
       }
     }
-    return this.#fn.call(null, input);
+    return this.#fn.call(null, input, this.#additionalArg);
   };
 }

packages/core/src/utils/index.ts

@@ -1,5 +1,9 @@
 import type { JSONValue } from "../global";
 
+export const isPromise = <T>(obj: unknown): obj is Promise<T> => {
+  return obj != null && typeof obj === "object" && "then" in obj;
+};
+
 export const isAsyncIterable = (
   obj: unknown,
 ): obj is AsyncIterable<unknown> => {

packages/core/tests/agents.test.ts

@@ -1,4 +1,5 @@
-import { validateAgentParams } from "@llamaindex/core/agent";
+import { LLMAgent, validateAgentParams } from "@llamaindex/core/agent";
+import { MockLLM } from "@llamaindex/core/utils";
 import { expect, test } from "vitest";
 import { ZodError } from "zod";
 
@@ -33,3 +34,62 @@ test("validate agent params", () => {
     ],
   });
 });
+
+test("LLMAgent streaming: first chunk should be available immediately", async () => {
+  const responseMessage =
+    "This is a very long response message that should take a while to stream";
+  const timeBetweenToken = 20; // delay time between tokens
+
+  const agent = new LLMAgent({
+    tools: [],
+    llm: new MockLLM({ responseMessage, timeBetweenToken }),
+  });
+
+  const startTime = Date.now();
+  const stream = await agent.chat({ message: "Hello", stream: true });
+
+  let fullResponse = "";
+  let timeToGetFirstChunk: number | undefined;
+
+  for await (const chunk of stream) {
+    expect(chunk).toHaveProperty("delta");
+    fullResponse += chunk.delta;
+    if (timeToGetFirstChunk === undefined) {
+      timeToGetFirstChunk = Date.now() - startTime;
+    }
+  }
+
+  expect(fullResponse).toBe(responseMessage);
+
+  // the first chunk should be available immediately and no need the whole response to be sent
+  expect(timeToGetFirstChunk).toBeLessThan(500);
+});
+
+test("LLMAgent create task: first task should be executed immediately", async () => {
+  const responseMessage =
+    "This is a very long response message that should take a while to stream";
+  const timeBetweenToken = 20; // delay time between tokens
+
+  const agent = new LLMAgent({
+    tools: [],
+    llm: new MockLLM({ responseMessage, timeBetweenToken }),
+  });
+
+  const startTime = Date.now();
+  const task = agent.createTask("Write a long paragraph", true, false, []);
+
+  let timeToGetFirstChunk: number | undefined;
+  let output: ReadableStream | undefined;
+  for await (const stepOutput of task) {
+    if (timeToGetFirstChunk === undefined) {
+      timeToGetFirstChunk = Date.now() - startTime;
+    }
+    if (stepOutput.output instanceof ReadableStream) {
+      output = stepOutput.output;
+    }
+  }
+
+  expect(timeToGetFirstChunk).toBeLessThan(500);
+  expect(output).toBeDefined();
+  expect(output).toBeInstanceOf(ReadableStream);
+});

packages/core/tests/node-parser/node-parser.test.ts

@@ -0,0 +1,24 @@
+import { NodeParser } from "@llamaindex/core/node-parser";
+import { TextNode } from "@llamaindex/core/schema";
+import { describe, expect, test } from "vitest";
+
+describe("NodeParser", () => {
+  test("node parser should allow async parse function", async () => {
+    class MyNodeParser extends NodeParser<Promise<TextNode[]>> {
+      protected async parseNodes(documents: TextNode[]): Promise<TextNode[]> {
+        await new Promise((resolve) => setTimeout(resolve, 1000));
+        return documents;
+      }
+    }
+
+    const nodeParser = new MyNodeParser();
+    const nodes = [
+      new TextNode({
+        text: "Hello, world!",
+      }),
+    ];
+    const result = nodeParser(nodes);
+    expect(result).toBeInstanceOf(Promise);
+    await expect(result).resolves.toEqual(nodes);
+  });
+});

packages/core/tests/tools.test.ts

@@ -1,5 +1,5 @@
 import { FunctionTool } from "@llamaindex/core/tools";
-import { describe, test } from "vitest";
+import { describe, expect, test, vi } from "vitest";
 import { z } from "zod";
 
 describe("FunctionTool", () => {
@@ -32,4 +32,38 @@ describe("FunctionTool", () => {
       parameters: inputSchema,
     });
   });
+
+  test("bind additional argument", () => {
+    type AdditionalHelloArgument = {
+      question?: string;
+    };
+
+    const hello = vi
+      .fn()
+      .mockImplementation((name: string, arg?: AdditionalHelloArgument) => {
+        return `Hello ${name}. ${arg?.question ?? ""}`;
+      });
+
+    const helloTool = FunctionTool.from<string, AdditionalHelloArgument>(
+      hello,
+      {
+        name: "hello",
+        description: "Says hello",
+      },
+    );
+
+    helloTool.call("Alice");
+    expect(hello).to.toHaveBeenCalledOnce();
+    expect(hello).to.toHaveBeenCalledWith("Alice", undefined);
+
+    hello.mockReset();
+
+    const additionalArg = {
+      question: "How is it going?",
+    };
+    const helloBoundTool = helloTool.bind(additionalArg);
+    helloBoundTool.call("Bob");
+    expect(hello).to.toHaveBeenCalledOnce();
+    expect(hello).to.toHaveBeenCalledWith("Bob", additionalArg);
+  });
 });

packages/env/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @llamaindex/env
 
+## 0.1.29
+
+### Patch Changes
+
+- beb922b: Fix edge runtime builds by adding missing packages to env package. Make gpt-tokenizer optional for llamaindex to reduce package size.
+
 ## 0.1.28
 
 ### 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": "0.1.28",
+  "version": "0.1.29",
   "type": "module",
   "types": "dist/index.d.ts",
   "module": "dist/index.js",
@@ -120,31 +120,22 @@
     "@types/node": "^22.9.0",
     "@types/readable-stream": "^4.0.15",
     "bunchee": "6.3.4",
-    "gpt-tokenizer": "^2.6.2",
-    "pathe": "^1.1.2",
     "vitest": "^2.1.5"
   },
-  "peerDependencies": {
+  "dependencies": {
+    "pathe": "^1.1.2",
     "@aws-crypto/sha256-js": "^5.2.0",
+    "js-tiktoken": "^1.0.12"
+  },
+  "peerDependencies": {
     "@huggingface/transformers": "^3.0.2",
-    "gpt-tokenizer": "^2.5.0",
-    "js-tiktoken": "^1.0.12",
-    "pathe": "^1.1.2"
+    "gpt-tokenizer": "^2.5.0"
   },
   "peerDependenciesMeta": {
-    "@aws-crypto/sha256-js": {
-      "optional": true
-    },
     "@huggingface/transformers": {
       "optional": true
     },
-    "pathe": {
-      "optional": true
-    },
-    "tiktoken": {
-      "optional": true
-    },
-    "js-tiktoken": {
+    "gpt-tokenizer": {
       "optional": true
     }
   }

packages/env/src/internal/tokenizers/node.ts

@@ -1,20 +1,48 @@
+import { getEncoding } from "js-tiktoken";
 import type { Tokenizer } from "./types.js";
 import { Tokenizers } from "./types.js";
 
-import cl100kBase from "gpt-tokenizer";
+function tryLoadGptTokenizer() {
+  try {
+    // eslint-disable-next-line @typescript-eslint/no-require-imports
+    return require("gpt-tokenizer"); // using require for CommonJS compatibility
+  } catch (e) {
+    return null;
+  }
+}
+
+const gptTokenizerModule = tryLoadGptTokenizer();
 
 class TokenizerSingleton {
   #defaultTokenizer: Tokenizer;
 
   constructor() {
-    this.#defaultTokenizer = {
-      encode: (text: string): Uint32Array => {
-        return new Uint32Array(cl100kBase.encode(text));
-      },
-      decode: (tokens: Uint32Array) => {
-        return cl100kBase.decode(tokens);
-      },
-    };
+    // Use gpt-tokenizer if available, otherwise use js-tiktoken
+    if (gptTokenizerModule) {
+      this.#defaultTokenizer = {
+        encode: (text: string): Uint32Array => {
+          return new Uint32Array(gptTokenizerModule.encode(text));
+        },
+        decode: (tokens: Uint32Array): string => {
+          return gptTokenizerModule.decode(Array.from(tokens));
+        },
+      };
+    } else {
+      // Fall back to js-tiktoken which is always available
+      // Note: js-tiktoken it's 60x slower than gpt-tokenizer
+      const encoding = getEncoding("cl100k_base");
+      this.#defaultTokenizer = {
+        encode: (text: string) => {
+          return new Uint32Array(encoding.encode(text));
+        },
+        decode: (tokens: Uint32Array) => {
+          const numberArray = Array.from(tokens);
+          const text = encoding.decode(numberArray);
+          const uint8Array = new TextEncoder().encode(text);
+          return new TextDecoder().decode(uint8Array);
+        },
+      };
+    }
   }
 
   tokenizer(encoding?: Tokenizers): Tokenizer {

packages/experimental/CHANGELOG.md

@@ -1,5 +1,44 @@
 # @llamaindex/experimental
 
+## 0.0.159
+
+### Patch Changes
+
+- llamaindex@0.9.9
+
+## 0.0.158
+
+### Patch Changes
+
+- Updated dependencies [bbc8c87]
+  - llamaindex@0.9.8
+
+## 0.0.157
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - llamaindex@0.9.7
+
+## 0.0.156
+
+### Patch Changes
+
+- llamaindex@0.9.6
+
+## 0.0.155
+
+### Patch Changes
+
+- llamaindex@0.9.5
+
+## 0.0.154
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - llamaindex@0.9.4
+
 ## 0.0.153
 
 ### Patch Changes

packages/experimental/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/experimental",
   "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.153",
+  "version": "0.0.159",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",
@@ -55,12 +55,10 @@
     "dev": "concurrently \"pnpm run build:esm --watch\" \"pnpm run build:cjs --watch\" \"pnpm run build:type --watch\""
   },
   "devDependencies": {
-    "@aws-crypto/sha256-js": "^5.2.0",
     "@swc/cli": "^0.5.0",
     "@swc/core": "^1.9.2",
     "@types/jsonpath": "^0.2.4",
-    "concurrently": "^9.1.0",
-    "pathe": "^1.1.2"
+    "concurrently": "^9.1.0"
   },
   "dependencies": {
     "@types/lodash": "^4.17.7",

packages/llamaindex/CHANGELOG.md

@@ -1,5 +1,70 @@
 # llamaindex
 
+## 0.9.9
+
+### Patch Changes
+
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+  - @llamaindex/cloud@3.0.8
+  - @llamaindex/node-parser@1.0.7
+  - @llamaindex/openai@0.1.59
+  - @llamaindex/workflow@0.0.15
+
+## 0.9.8
+
+### Patch Changes
+
+- bbc8c87: fix: prefer using embedding model from vector store
+- Updated dependencies [4b49428]
+  - @llamaindex/workflow@0.0.14
+
+## 0.9.7
+
+### Patch Changes
+
+- beb922b: Fix edge runtime builds by adding missing packages to env package. Make gpt-tokenizer optional for llamaindex to reduce package size.
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+  - @llamaindex/cloud@3.0.7
+  - @llamaindex/node-parser@1.0.6
+  - @llamaindex/openai@0.1.58
+  - @llamaindex/workflow@0.0.13
+
+## 0.9.6
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+  - @llamaindex/workflow@0.0.12
+  - @llamaindex/cloud@3.0.6
+  - @llamaindex/node-parser@1.0.5
+  - @llamaindex/openai@0.1.57
+
+## 0.9.5
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+  - @llamaindex/cloud@3.0.5
+  - @llamaindex/node-parser@1.0.4
+  - @llamaindex/openai@0.1.56
+
+## 0.9.4
+
+### Patch Changes
+
+- cb021e7: feat(node-parser): support async function
+- Updated dependencies [cb256f2]
+- Updated dependencies [cb021e7]
+  - @llamaindex/openai@0.1.55
+  - @llamaindex/core@0.5.3
+  - @llamaindex/cloud@3.0.4
+  - @llamaindex/node-parser@1.0.3
+
 ## 0.9.3
 
 ### Patch Changes

packages/llamaindex/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llamaindex",
-  "version": "0.9.3",
+  "version": "0.9.9",
   "license": "MIT",
   "type": "module",
   "keywords": [
@@ -25,12 +25,12 @@
     "@llamaindex/env": "workspace:*",
     "@llamaindex/node-parser": "workspace:*",
     "@llamaindex/openai": "workspace:*",
+    "@llamaindex/workflow": "workspace:*",
     "@types/lodash": "^4.17.7",
     "@types/node": "^22.9.0",
     "ajv": "^8.17.1",
     "lodash": "^4.17.21",
-    "magic-bytes.js": "^1.10.0",
-    "gpt-tokenizer": "^2.6.2"
+    "magic-bytes.js": "^1.10.0"
   },
   "devDependencies": {
     "@swc/cli": "^0.5.0",

packages/llamaindex/src/embeddings/OpenAIEmbedding.ts

@@ -1 +0,0 @@
-export * from "@llamaindex/openai";

packages/llamaindex/src/embeddings/index.ts

@@ -1,5 +0,0 @@
-export * from "@llamaindex/core/embeddings";
-export { FireworksEmbedding } from "./fireworks.js";
-export * from "./JinaAIEmbedding.js";
-export * from "./OpenAIEmbedding.js";
-export { TogetherEmbedding } from "./together.js";

packages/llamaindex/src/index.edge.ts

@@ -22,6 +22,7 @@ export {
 export * from "@llamaindex/core/agent";
 export * from "@llamaindex/core/chat-engine";
 export * from "@llamaindex/core/data-structs";
+export * from "@llamaindex/core/embeddings";
 export {
   CallbackManager,
   DEFAULT_BASE_URL,
@@ -65,9 +66,10 @@ export * from "@llamaindex/core/storage/doc-store";
 export * from "@llamaindex/core/storage/index-store";
 export * from "@llamaindex/core/storage/kv-store";
 export * from "@llamaindex/core/utils";
+export * from "@llamaindex/openai";
+export * from "@llamaindex/workflow/agent";
 export * from "./agent/index.js";
 export * from "./cloud/index.js";
-export * from "./embeddings/index.js";
 export * from "./engines/chat/index.js";
 export * from "./engines/query/index.js";
 export * from "./evaluation/index.js";
@@ -75,7 +77,6 @@ export * from "./extractors/index.js";
 export * from "./indices/index.js";
 export * from "./ingestion/index.js";
 export { imageToDataUrl } from "./internal/utils.js";
-export * from "./llm/index.js";
 export * from "./node-parser.js";
 export * from "./objects/index.js";
 export * from "./OutputParser.js";

packages/llamaindex/src/index.ts

@@ -1,8 +1,5 @@
 export * from "./index.edge.js";
 
-// TODO: clean up, move to jinaai package
-export { JinaAIEmbedding } from "./embeddings/JinaAIEmbedding.js";
-
 // Don't export file-system stores for non-node.js runtime on top level,
 //  as we cannot guarantee that they will work in other environments
 export * from "./storage/index.js";

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

@@ -296,7 +296,7 @@ export class KeywordTableIndex extends BaseIndex<KeywordTable> {
       await docStore.setDocumentHash(doc.id_, doc.hash);
     }
 
-    const nodes = Settings.nodeParser.getNodesFromDocuments(documents);
+    const nodes = await Settings.nodeParser.getNodesFromDocuments(documents);
     const index = await KeywordTableIndex.init({
       nodes,
       storageContext,

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

@@ -145,7 +145,7 @@ export class SummaryIndex extends BaseIndex<IndexList> {
       await docStore.setDocumentHash(doc.id_, doc.hash);
     }
 
-    const nodes = Settings.nodeParser.getNodesFromDocuments(documents);
+    const nodes = await Settings.nodeParser.getNodesFromDocuments(documents);
 
     const index = await SummaryIndex.init({
       nodes,

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

@@ -175,7 +175,7 @@ export class VectorStoreIndex extends BaseIndex<IndexDict> {
     for (const type in nodeMap) {
       const nodes = nodeMap[type as ModalityType];
       const embedModel =
-        this.embedModel ?? this.vectorStores[type as ModalityType]?.embedModel;
+        this.vectorStores[type as ModalityType]?.embedModel ?? this.embedModel;
       if (embedModel && nodes) {
         await embedModel(nodes, {
           logProgress: options?.logProgress,

packages/llamaindex/src/llm/index.ts

@@ -1,4 +0,0 @@
-export { DeepSeekLLM } from "./deepseek.js";
-export { FireworksLLM } from "./fireworks.js";
-export * from "./openai.js";
-export { TogetherLLM } from "./together.js";

packages/llamaindex/src/llm/openai.ts

@@ -1 +0,0 @@
-export * from "@llamaindex/openai";

packages/llamaindex/tests/MetadataExtractors.test.ts

@@ -1,13 +1,11 @@
 import { Document } from "@llamaindex/core/schema";
-import { Settings } from "llamaindex";
-import { OpenAIEmbedding } from "llamaindex/embeddings/index";
+import { OpenAI, OpenAIEmbedding, Settings } from "llamaindex";
 import {
   KeywordExtractor,
   QuestionsAnsweredExtractor,
   SummaryExtractor,
   TitleExtractor,
 } from "llamaindex/extractors/index";
-import { OpenAI } from "llamaindex/llm/openai";
 import { SentenceSplitter } from "llamaindex/node-parser";
 import { afterAll, beforeAll, describe, expect, test, vi } from "vitest";
 import {

packages/llamaindex/tests/Selectors.test.ts

@@ -1,7 +1,7 @@
 import { describe, expect, test } from "vitest";
 // from unittest.mock import patch
 
-import { OpenAI } from "llamaindex/llm/index";
+import { OpenAI } from "llamaindex";
 import { LLMSingleSelector } from "llamaindex/selectors/index";
 import { mocStructuredkLlmGeneration } from "./utility/mockOpenAI.js";
 

packages/llamaindex/tests/indices/SummaryIndex.test.ts

@@ -20,13 +20,13 @@ describe("SummaryIndex", () => {
   let storageContext: StorageContext;
 
   beforeAll(async () => {
-    storageContext = await storageContextFromDefaults({
-      persistDir: testDir,
-    });
-
     const embedModel = new OpenAIEmbedding();
     mockEmbeddingModel(embedModel);
     Settings.embedModel = embedModel;
+
+    storageContext = await storageContextFromDefaults({
+      persistDir: testDir,
+    });
   });
 
   afterAll(() => {

packages/llamaindex/tests/indices/VectorStoreIndex.test.ts

@@ -9,7 +9,7 @@ import { DocStoreStrategy } from "llamaindex/ingestion/strategies/index";
 import { mkdtemp, rm } from "node:fs/promises";
 import { tmpdir } from "node:os";
 import { join } from "node:path";
-import { afterAll, beforeAll, describe, expect, test, vi } from "vitest";
+import { afterAll, beforeAll, describe, expect, it, test, vi } from "vitest";
 
 const testDir = await mkdtemp(join(tmpdir(), "test-"));
 
@@ -24,6 +24,10 @@ describe("VectorStoreIndex", () => {
   ) => Promise<Array<number>>;
 
   beforeAll(async () => {
+    const embedModel = new OpenAIEmbedding();
+    mockEmbeddingModel(embedModel);
+    Settings.embedModel = embedModel;
+
     storageContext = await mockStorageContext(testDir);
     testStrategy = async (
       strategy: DocStoreStrategy,
@@ -41,10 +45,6 @@ describe("VectorStoreIndex", () => {
       }
       return entries;
     };
-
-    const embedModel = new OpenAIEmbedding();
-    mockEmbeddingModel(embedModel);
-    Settings.embedModel = embedModel;
   });
 
   afterAll(() => {
@@ -65,3 +65,28 @@ describe("VectorStoreIndex", () => {
     await rm(testDir, { recursive: true });
   });
 });
+
+describe("[VectorStoreIndex] use embedding model", () => {
+  it("should use embedding model passed in options instead of Settings", async () => {
+    const documents = [new Document({ text: "This needs to be embedded" })];
+
+    // Create mock embedding models
+    const settingsEmbedModel = new OpenAIEmbedding();
+    const customEmbedModel = new OpenAIEmbedding();
+
+    // Mock the embedding models using the utility function
+    mockEmbeddingModel(settingsEmbedModel);
+    mockEmbeddingModel(customEmbedModel);
+
+    // Add spies to track calls
+    const settingsSpy = vi.spyOn(settingsEmbedModel, "getTextEmbeddings");
+    const customSpy = vi.spyOn(customEmbedModel, "getTextEmbeddings");
+
+    Settings.embedModel = settingsEmbedModel;
+
+    const storageContext = await mockStorageContext(testDir, customEmbedModel); // setup custom embedding model
+    await VectorStoreIndex.fromDocuments(documents, { storageContext });
+    expect(customSpy).toHaveBeenCalled();
+    expect(settingsSpy).not.toHaveBeenCalled();
+  });
+});

packages/llamaindex/tests/utility/mockOpenAI.ts

@@ -1,8 +1,6 @@
 import type { CallbackManager } from "@llamaindex/core/global";
-import type { LLMChatParamsBase } from "llamaindex";
-import { Settings } from "llamaindex";
-import type { OpenAIEmbedding } from "llamaindex/embeddings/OpenAIEmbedding";
-import { OpenAI } from "llamaindex/llm/openai";
+import type { LLMChatParamsBase, OpenAIEmbedding } from "llamaindex";
+import { OpenAI, Settings } from "llamaindex";
 import { vi } from "vitest";
 
 export const DEFAULT_LLM_TEXT_OUTPUT = "MOCK_TOKEN_1-MOCK_TOKEN_2";

packages/llamaindex/tests/utility/mockStorageContext.ts

@@ -1,14 +1,27 @@
-import { OpenAIEmbedding, storageContextFromDefaults } from "llamaindex";
+import {
+  BaseEmbedding,
+  OpenAIEmbedding,
+  storageContextFromDefaults,
+} from "llamaindex";
 
 import { mockEmbeddingModel } from "./mockOpenAI.js";
 
-export async function mockStorageContext(testDir: string) {
+export async function mockStorageContext(
+  testDir: string,
+  embeddingModel?: BaseEmbedding,
+) {
   const storageContext = await storageContextFromDefaults({
     persistDir: testDir,
   });
   for (const store of Object.values(storageContext.vectorStores)) {
-    store.embedModel = new OpenAIEmbedding();
-    mockEmbeddingModel(store.embedModel as OpenAIEmbedding);
+    if (embeddingModel) {
+      // use embeddingModel if it is passed in
+      store.embedModel = embeddingModel;
+    } else {
+      // mock an embedding model for testing
+      store.embedModel = new OpenAIEmbedding();
+      mockEmbeddingModel(store.embedModel as OpenAIEmbedding);
+    }
   }
   return storageContext;
 }

packages/node-parser/CHANGELOG.md

@@ -1,5 +1,41 @@
 # @llamaindex/node-parser
 
+## 1.0.7
+
+### Patch Changes
+
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+
+## 1.0.6
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+
+## 1.0.5
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+
+## 1.0.4
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+
+## 1.0.3
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - @llamaindex/core@0.5.3
+
 ## 1.0.2
 
 ### Patch Changes

packages/node-parser/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/node-parser",
-  "version": "1.0.2",
+  "version": "1.0.7",
   "description": "Node parser for LlamaIndex",
   "type": "module",
   "exports": {

packages/providers/anthropic/CHANGELOG.md

@@ -1,5 +1,42 @@
 # @llamaindex/anthropic
 
+## 0.2.5
+
+### Patch Changes
+
+- 8bf1ca1: Support chat stream with tools for Anthropic LLM
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+
+## 0.2.4
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+
+## 0.2.3
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+
+## 0.2.2
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+
+## 0.2.1
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - @llamaindex/core@0.5.3
+
 ## 0.2.0
 
 ### Minor Changes

packages/providers/anthropic/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/anthropic",
   "description": "Anthropic Adapter for LlamaIndex",
-  "version": "0.2.0",
+  "version": "0.2.5",
   "type": "module",
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",

packages/providers/anthropic/src/llm.ts

@@ -13,7 +13,7 @@ import type {
   Tool,
   ToolUseBlock,
 } from "@anthropic-ai/sdk/resources/messages";
-import { wrapLLMEvent } from "@llamaindex/core/decorator";
+import { wrapEventCaller, wrapLLMEvent } from "@llamaindex/core/decorator";
 import type { JSONObject } from "@llamaindex/core/global";
 import type {
   BaseTool,
@@ -22,6 +22,7 @@ import type {
   ChatResponseChunk,
   LLMChatParamsNonStreaming,
   LLMChatParamsStreaming,
+  PartialToolCall,
   ToolCallLLMMessageOptions,
 } from "@llamaindex/core/llms";
 import { ToolCallLLM } from "@llamaindex/core/llms";
@@ -378,6 +379,7 @@ export class Anthropic extends ToolCallLLM<
       AnthropicToolCallLLMMessageOptions
     >,
   ): Promise<ChatResponse<AnthropicToolCallLLMMessageOptions>>;
+  @wrapEventCaller
   @wrapLLMEvent
   async chat(
     params:
@@ -445,19 +447,16 @@ export class Anthropic extends ToolCallLLM<
       ),
     };
 
-    if (stream) {
-      if (tools) {
-        console.error("Tools are not supported in streaming mode");
-      }
-      return this.streamChat(anthropic, apiParams);
-    }
-
     if (tools?.length) {
       Object.assign(apiParams, {
         tools: this.prepareToolsForAPI(tools),
       });
     }
 
+    if (stream) {
+      return this.streamChat(anthropic, apiParams);
+    }
+
     const response = await anthropic.messages.create(apiParams);
 
     const toolUseBlock = response.content.filter(
@@ -494,6 +493,7 @@ export class Anthropic extends ToolCallLLM<
     };
   }
 
+  @wrapEventCaller
   protected async *streamChat(
     anthropic: SDKAnthropic,
     params: MessageCreateParams,
@@ -503,7 +503,9 @@ export class Anthropic extends ToolCallLLM<
       stream: true,
     });
 
-    let idx_counter: number = 0;
+    let currentToolCall: PartialToolCall | null = null;
+    let accumulatedToolInput = "";
+
     for await (const part of stream) {
       const textContent =
         part.type === "content_block_delta" && part.delta.type === "text_delta"
@@ -516,9 +518,48 @@ export class Anthropic extends ToolCallLLM<
           ? part.delta.thinking
           : undefined;
 
+      if (
+        part.type === "content_block_start" &&
+        part.content_block.type === "tool_use"
+      ) {
+        currentToolCall = {
+          id: part.content_block.id,
+          name: part.content_block.name,
+          input: "",
+        };
+        accumulatedToolInput = "";
+        continue;
+      }
+
+      if (
+        part.type === "content_block_delta" &&
+        part.delta.type === "input_json_delta" &&
+        currentToolCall
+      ) {
+        accumulatedToolInput += part.delta.partial_json;
+        continue;
+      }
+
+      if (part.type === "content_block_stop" && currentToolCall) {
+        yield {
+          raw: part,
+          delta: "",
+          options: {
+            toolCall: [
+              {
+                id: currentToolCall.id,
+                name: currentToolCall.name,
+                input: accumulatedToolInput,
+              },
+            ],
+          },
+        };
+        currentToolCall = null;
+        continue;
+      }
+
       if (!textContent && !thinking) continue;
 
-      idx_counter++;
       yield {
         raw: part,
         delta: textContent ?? "",

packages/providers/clip/CHANGELOG.md

@@ -1,5 +1,47 @@
 # @llamaindex/clip
 
+## 0.0.43
+
+### Patch Changes
+
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+  - @llamaindex/openai@0.1.59
+
+## 0.0.42
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+  - @llamaindex/openai@0.1.58
+
+## 0.0.41
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+  - @llamaindex/openai@0.1.57
+
+## 0.0.40
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+  - @llamaindex/openai@0.1.56
+
+## 0.0.39
+
+### Patch Changes
+
+- Updated dependencies [cb256f2]
+- Updated dependencies [cb021e7]
+  - @llamaindex/openai@0.1.55
+  - @llamaindex/core@0.5.3
+
 ## 0.0.38
 
 ### Patch Changes

packages/providers/clip/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/clip",
   "description": "Clip Embedding Adapter for LlamaIndex",
-  "version": "0.0.38",
+  "version": "0.0.43",
   "type": "module",
   "types": "dist/index.d.ts",
   "main": "dist/index.cjs",

packages/providers/cohere/CHANGELOG.md

@@ -1,5 +1,41 @@
 # @llamaindex/cohere
 
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [4bac71d]
+  - @llamaindex/core@0.5.7
+
+## 0.0.11
+
+### Patch Changes
+
+- Updated dependencies [beb922b]
+  - @llamaindex/env@0.1.29
+  - @llamaindex/core@0.5.6
+
+## 0.0.10
+
+### Patch Changes
+
+- Updated dependencies [5668970]
+  - @llamaindex/core@0.5.5
+
+## 0.0.9
+
+### Patch Changes
+
+- Updated dependencies [ad3c7f1]
+  - @llamaindex/core@0.5.4
+
+## 0.0.8
+
+### Patch Changes
+
+- Updated dependencies [cb021e7]
+  - @llamaindex/core@0.5.3
+
 ## 0.0.7
 
 ### Patch Changes

packages/providers/cohere/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/cohere",
   "description": "Cohere Adapter for LlamaIndex",
-  "version": "0.0.7",
+  "version": "0.0.12",
   "type": "module",
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",