Release 0.9.8 (#1711 )

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: marcusschiesser <17126+marcusschiesser@users.noreply.github.com>
fix: prefer using embedding model from vector store (#1708 )
2026-07-01 22:14:03 -04:00 · 2025-03-06 16:33:19 +07:00 · 2025-03-06 16:24:05 +07:00 · 2025-03-06 11:13:42 +07:00 · 2025-03-06 11:04:56 +07:00 · 2025-03-05 16:59:05 +07:00
203 changed files with 5923 additions and 377 deletions
@@ -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/*"
 ```

@@ -1,5 +1,77 @@
 # @llamaindex/doc

+## 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
+
+- Updated dependencies [d952e68]
+  - @llamaindex/core@0.5.2
+  - @llamaindex/cloud@3.0.3
+  - llamaindex@0.9.3
+  - @llamaindex/node-parser@1.0.2
+  - @llamaindex/openai@0.1.54
+  - @llamaindex/readers@2.0.2
+
 ## 0.1.2

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/doc",
-  "version": "0.1.2",
+  "version": "0.1.8",
  "private": true,
  "scripts": {
    "build": "pnpm run build:docs && next build",
@@ -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
@@ -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.
+
@@ -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();

@@ -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>",
@@ -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)
@@ -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.
@@ -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>",
@@ -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>",
@@ -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>",
@@ -1,5 +1,44 @@
 # @llamaindex/cloudflare-worker-agent-test

+## 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
+
+- llamaindex@0.9.3
+
 ## 0.0.136

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.136",
+  "version": "0.0.142",
  "type": "module",
  "private": true,
  "scripts": {
@@ -1,5 +1,35 @@
 # @llamaindex/llama-parse-browser-test

+## 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
+
+- @llamaindex/cloud@3.0.3
+
 ## 0.0.47

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/llama-parse-browser-test",
  "private": true,
-  "version": "0.0.47",
+  "version": "0.0.52",
  "type": "module",
  "scripts": {
    "dev": "vite",
@@ -1,5 +1,44 @@
 # @llamaindex/next-agent-test

+## 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
+
+- llamaindex@0.9.3
+
 ## 0.1.136

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/next-agent-test",
-  "version": "0.1.136",
+  "version": "0.1.142",
  "private": true,
  "scripts": {
    "dev": "next dev",
@@ -1,5 +1,44 @@
 # test-edge-runtime

+## 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
+
+- llamaindex@0.9.3
+
 ## 0.1.135

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.135",
+  "version": "0.1.141",
  "private": true,
  "scripts": {
    "dev": "next dev",
@@ -1,5 +1,54 @@
 # @llamaindex/next-node-runtime

+## 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
+
+- llamaindex@0.9.3
+- @llamaindex/huggingface@0.0.38
+- @llamaindex/readers@2.0.2
+
 ## 0.1.2

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.1.2",
+  "version": "0.1.8",
  "private": true,
  "scripts": {
    "dev": "next dev",
@@ -1,5 +1,44 @@
 # vite-import-llamaindex

+## 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
+
+- llamaindex@0.9.3
+
 ## 0.0.2

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "vite-import-llamaindex",
  "private": true,
-  "version": "0.0.2",
+  "version": "0.0.8",
  "type": "module",
  "scripts": {
    "build": "vite build",
@@ -1,5 +1,45 @@
 # @llamaindex/waku-query-engine-test

+## 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
+
+- llamaindex@0.9.3
+
 ## 0.0.136

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.136",
+  "version": "0.0.142",
  "type": "module",
  "private": true,
  "scripts": {
@@ -1,5 +1,221 @@
 # examples

+## 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
+
+- 3564244: Add support for claude 3.7 thinking (and set default temperature to 1)
+- 3564244: add support for claude 3.7
+- Updated dependencies [d952e68]
+- Updated dependencies [a9c6144]
+- Updated dependencies [5c026e8]
+- Updated dependencies [3564244]
+- Updated dependencies [3564244]
+  - @llamaindex/core@0.5.2
+  - @llamaindex/anthropic@0.2.0
+  - @llamaindex/firestore@1.0.0
+  - @llamaindex/cloud@3.0.3
+  - llamaindex@0.9.3
+  - @llamaindex/node-parser@1.0.2
+  - @llamaindex/clip@0.0.38
+  - @llamaindex/cohere@0.0.7
+  - @llamaindex/deepinfra@0.0.38
+  - @llamaindex/google@0.0.9
+  - @llamaindex/huggingface@0.0.38
+  - @llamaindex/mistral@0.0.7
+  - @llamaindex/mixedbread@0.0.7
+  - @llamaindex/ollama@0.0.42
+  - @llamaindex/openai@0.1.54
+  - @llamaindex/portkey-ai@0.0.35
+  - @llamaindex/replicate@0.0.35
+  - @llamaindex/astra@0.0.7
+  - @llamaindex/azure@0.1.2
+  - @llamaindex/chroma@0.0.7
+  - @llamaindex/milvus@0.1.2
+  - @llamaindex/mongodb@0.0.7
+  - @llamaindex/pinecone@0.0.7
+  - @llamaindex/postgres@0.0.35
+  - @llamaindex/qdrant@0.1.2
+  - @llamaindex/upstash@0.0.7
+  - @llamaindex/weaviate@0.0.7
+  - @llamaindex/vercel@0.0.13
+  - @llamaindex/readers@2.0.2
+  - @llamaindex/groq@0.0.53
+  - @llamaindex/vllm@0.0.24
+
 ## 0.2.1

 ### Patch Changes
@@ -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);
+});
@@ -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);
+});
@@ -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);
+});
@@ -8,7 +8,7 @@ Settings.callbackManager.on("llm-tool-call", (event) => {

 const anthropic = new Anthropic({
  apiKey: process.env.ANTHROPIC_API_KEY,
-  model: "claude-3-5-sonnet",
+  model: "claude-3-7-sonnet",
 });

 const agent = new AnthropicAgent({
@@ -3,7 +3,7 @@ import { Anthropic } from "@llamaindex/anthropic";
 (async () => {
  const anthropic = new Anthropic({
    apiKey: process.env.ANTHROPIC_API_KEY,
-    model: "claude-3-opus",
+    model: "claude-3-7-sonnet",
  });
  const result = await anthropic.chat({
    messages: [
@@ -15,5 +15,5 @@ import { Anthropic } from "@llamaindex/anthropic";
      },
    ],
  });
-  console.log(result);
+  console.log(result.message.content);
 })();
@@ -6,7 +6,7 @@ import readline from "node:readline/promises";
 (async () => {
  const llm = new Anthropic({
    apiKey: process.env.ANTHROPIC_API_KEY,
-    model: "claude-3-opus",
+    model: "claude-3-7-sonnet",
  });
  // chatHistory will store all the messages in the conversation
  const chatHistory = new ChatMemoryBuffer({
@@ -2,7 +2,7 @@ import { Anthropic } from "@llamaindex/anthropic";

 async function main() {
  const anthropic = new Anthropic({
-    model: "claude-3-5-sonnet-20241022",
+    model: "claude-3-7-sonnet",
  });

  const entireBook = await fetch(
@@ -3,7 +3,7 @@ import { Anthropic } from "@llamaindex/anthropic";
 (async () => {
  const anthropic = new Anthropic({
    apiKey: process.env.ANTHROPIC_API_KEY,
-    model: "claude-3-5-sonnet",
+    model: "claude-3-7-sonnet",
  });
  const stream = await anthropic.chat({
    messages: [
@@ -0,0 +1,32 @@
+import { Anthropic } from "@llamaindex/anthropic";
+
+(async () => {
+  const anthropic = new Anthropic({
+    model: "claude-3-7-sonnet",
+    maxTokens: 20000,
+    additionalChatOptions: {
+      thinking: {
+        type: "enabled",
+        budget_tokens: 16000,
+      },
+    },
+  });
+  const result = await anthropic.chat({
+    messages: [
+      {
+        role: "user",
+        content:
+          "Are there an infinite number of prime numbers such that n mod 4 == 3?",
+      },
+    ],
+    stream: true,
+  });
+  console.log("Thinking...");
+  for await (const chunk of result) {
+    if (chunk.delta) {
+      process.stdout.write(chunk.delta);
+    } else if (chunk.options?.thinking) {
+      process.stdout.write(chunk.options.thinking);
+    }
+  }
+})();
@@ -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");
+  }
+})();
@@ -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");
+})();
@@ -0,0 +1,4 @@
+GCP_PROJECT_ID=
+GCP_CREDENTIALS=
+FIRESTORE_DB=
+OPENAI_API_KEY=
@@ -0,0 +1,35 @@
+# Firestore Vector Store
+
+Here are some sample scripts which work with loading and querying data from a Firestore Vector Store.
+
+## Prerequisites
+
+- A Firestore Database
+  - Hosted https://console.firebase.google.com/
+- An OpenAI API Key
+
+## Setup
+
+1. Set your env variables:
+
+- `FIRESTORE_DB`: Name of your Firestore database
+- `GCP_PROJECT_ID`: Your GCP project ID
+- `GCP_CREDENTIALS`: Your GCP credentials JSON
+- `OPENAI_API_KEY`: Your OpenAI key
+
+2. `cd` Into the `examples` directory
+3. run `npm i`
+
+## Load the data
+
+This sample loads the same dataset of movie reviews as sample dataset
+
+run `npx tsx firestore/load.ts`
+
+## Use RAG to Query the data
+
+run `npx tsx firestore/query.ts`
+
+## Delete the data
+
+run `npx tsx firestore/delete.ts`
@@ -0,0 +1,32 @@
+import { CollectionReference } from "@google-cloud/firestore";
+import "dotenv/config";
+
+import { FirestoreVectorStore } from "@llamaindex/firestore";
+import { OpenAIEmbedding, Settings } from "llamaindex";
+
+const indexName = "MovieReviews";
+
+Settings.embedModel = new OpenAIEmbedding();
+
+async function main() {
+  try {
+    const vectorStore = new FirestoreVectorStore({
+      clientOptions: {
+        credentials: JSON.parse(process.env.GCP_CREDENTIALS!),
+        projectId: process.env.GCP_PROJECT_ID!,
+        databaseId: process.env.FIRESTORE_DB!,
+        ignoreUndefinedProperties: true,
+      },
+      collectionName: indexName,
+      customCollectionReference: (rootCollection: CollectionReference) => {
+        return rootCollection.doc("accountId-123").collection("vectors");
+      },
+    });
+
+    vectorStore.delete("movie_reviews.csv");
+  } catch (e) {
+    console.error(e);
+  }
+}
+
+void main();
@@ -0,0 +1,44 @@
+import { CollectionReference } from "@google-cloud/firestore";
+import { CSVReader } from "@llamaindex/readers/csv";
+import "dotenv/config";
+
+import {
+  OpenAIEmbedding,
+  Settings,
+  storageContextFromDefaults,
+  VectorStoreIndex,
+} from "llamaindex";
+
+import { FirestoreVectorStore } from "@llamaindex/firestore";
+
+const indexName = "MovieReviews";
+
+Settings.embedModel = new OpenAIEmbedding();
+
+async function main() {
+  try {
+    const reader = new CSVReader(false);
+    const docs = await reader.loadData("./data/movie_reviews.csv");
+
+    const vectorStore = new FirestoreVectorStore({
+      clientOptions: {
+        credentials: JSON.parse(process.env.GCP_CREDENTIALS!),
+        projectId: process.env.GCP_PROJECT_ID!,
+        databaseId: process.env.FIRESTORE_DB!,
+        ignoreUndefinedProperties: true,
+      },
+      collectionName: indexName,
+      customCollectionReference: (rootCollection: CollectionReference) => {
+        return rootCollection.doc("accountId-123").collection("vectors");
+      },
+    });
+
+    const storageContext = await storageContextFromDefaults({ vectorStore });
+
+    await VectorStoreIndex.fromDocuments(docs, { storageContext });
+  } catch (e) {
+    console.error(e);
+  }
+}
+
+void main();
@@ -0,0 +1,58 @@
+import "dotenv/config";
+
+import { OpenAIEmbedding, Settings, VectorStoreIndex } from "llamaindex";
+
+import { CollectionReference } from "@google-cloud/firestore";
+import { FirestoreVectorStore } from "@llamaindex/firestore";
+
+const indexName = "MovieReviews";
+
+Settings.embedModel = new OpenAIEmbedding();
+
+async function main() {
+  try {
+    const vectorStore = new FirestoreVectorStore({
+      clientOptions: {
+        credentials: JSON.parse(process.env.GCP_CREDENTIALS!),
+        projectId: process.env.GCP_PROJECT_ID!,
+        databaseId: process.env.FIRESTORE_DB!,
+        ignoreUndefinedProperties: true,
+      },
+      collectionName: indexName,
+      customCollectionReference: (rootCollection: CollectionReference) => {
+        return rootCollection.doc("accountId-123").collection("vectors");
+      },
+    });
+    const index = await VectorStoreIndex.fromVectorStore(vectorStore);
+    const retriever = index.asRetriever({ similarityTopK: 20 });
+
+    const queryEngine = index.asQueryEngine({ retriever });
+    const query = "Get all movie titles.";
+    const results = await queryEngine.query({ query });
+    console.log(`Query from ${results.sourceNodes?.length} nodes`);
+    console.log(results.response);
+
+    console.log("\n=====\nQuerying the index with filters");
+    const queryEngineWithFilters = index.asQueryEngine({
+      retriever,
+      preFilters: {
+        filters: [
+          {
+            key: "file_name",
+            value: "movie_reviews.csv",
+            operator: "==",
+          },
+        ],
+      },
+    });
+    const resultAfterFilter = await queryEngineWithFilters.query({
+      query: "Get all movie titles.",
+    });
+    console.log(`Query from ${resultAfterFilter.sourceNodes?.length} nodes`);
+    console.log(resultAfterFilter.response);
+  } catch (e) {
+    console.error(e);
+  }
+}
+
+void main();
@@ -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() {
@@ -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?" });
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/examples",
-  "version": "0.2.1",
+  "version": "0.2.7",
  "private": true,
  "scripts": {
    "lint": "eslint .",
@@ -11,37 +11,43 @@
    "@azure/cosmos": "^4.1.1",
    "@azure/identity": "^4.4.1",
    "@azure/search-documents": "^12.1.0",
-    "@llamaindex/anthropic": "^0.1.1",
-    "@llamaindex/astra": "^0.0.6",
-    "@llamaindex/azure": "^0.1.1",
-    "@llamaindex/chroma": "^0.0.6",
-    "@llamaindex/clip": "^0.0.37",
-    "@llamaindex/cloud": "^3.0.1",
-    "@llamaindex/cohere": "^0.0.6",
-    "@llamaindex/core": "^0.5.1",
-    "@llamaindex/deepinfra": "^0.0.37",
-    "@llamaindex/env": "^0.1.28",
-    "@llamaindex/google": "^0.0.8",
-    "@llamaindex/groq": "^0.0.52",
-    "@llamaindex/huggingface": "^0.0.37",
-    "@llamaindex/milvus": "^0.1.1",
-    "@llamaindex/mistral": "^0.0.6",
-    "@llamaindex/mixedbread": "^0.0.6",
-    "@llamaindex/mongodb": "^0.0.6",
-    "@llamaindex/node-parser": "^1.0.1",
-    "@llamaindex/ollama": "^0.0.41",
-    "@llamaindex/openai": "^0.1.53",
-    "@llamaindex/pinecone": "^0.0.6",
-    "@llamaindex/portkey-ai": "^0.0.34",
-    "@llamaindex/postgres": "^0.0.34",
-    "@llamaindex/qdrant": "^0.1.1",
-    "@llamaindex/readers": "^2.0.1",
-    "@llamaindex/replicate": "^0.0.34",
-    "@llamaindex/upstash": "^0.0.6",
-    "@llamaindex/vercel": "^0.0.12",
-    "@llamaindex/vllm": "^0.0.23",
-    "@llamaindex/weaviate": "^0.0.6",
-    "@llamaindex/workflow": "^0.0.11",
+    "@llamaindex/anthropic": "^0.2.4",
+    "@llamaindex/astra": "^0.0.11",
+    "@llamaindex/azure": "^0.1.6",
+    "@llamaindex/chroma": "^0.0.11",
+    "@llamaindex/clip": "^0.0.42",
+    "@llamaindex/cloud": "^3.0.7",
+    "@llamaindex/cohere": "^0.0.11",
+    "@llamaindex/core": "^0.5.6",
+    "@llamaindex/deepinfra": "^0.0.42",
+    "@llamaindex/env": "^0.1.29",
+    "@llamaindex/firestore": "^1.0.4",
+    "@llamaindex/google": "^0.0.14",
+    "@llamaindex/groq": "^0.0.57",
+    "@llamaindex/huggingface": "^0.0.42",
+    "@llamaindex/milvus": "^0.1.6",
+    "@llamaindex/mistral": "^0.0.11",
+    "@llamaindex/mixedbread": "^0.0.11",
+    "@llamaindex/mongodb": "^0.0.11",
+    "@llamaindex/node-parser": "^1.0.6",
+    "@llamaindex/ollama": "^0.0.46",
+    "@llamaindex/openai": "^0.1.58",
+    "@llamaindex/pinecone": "^0.0.11",
+    "@llamaindex/portkey-ai": "^0.0.39",
+    "@llamaindex/postgres": "^0.0.39",
+    "@llamaindex/qdrant": "^0.1.6",
+    "@llamaindex/readers": "^2.0.6",
+    "@llamaindex/replicate": "^0.0.39",
+    "@llamaindex/upstash": "^0.0.11",
+    "@llamaindex/vercel": "^0.0.17",
+    "@llamaindex/vllm": "^0.0.28",
+    "@llamaindex/voyage-ai": "^1.0.3",
+    "@llamaindex/weaviate": "^0.0.11",
+    "@llamaindex/workflow": "^0.0.14",
+    "@llamaindex/deepseek": "^0.0.2",
+    "@llamaindex/fireworks": "^0.0.2",
+    "@llamaindex/together": "^0.0.2",
+    "@llamaindex/jinaai": "^0.0.2",
    "@notionhq/client": "^2.2.15",
    "@pinecone-database/pinecone": "^4.0.0",
    "@vercel/postgres": "^0.10.0",
@@ -50,11 +56,11 @@
    "commander": "^12.1.0",
    "dotenv": "^16.4.5",
    "js-tiktoken": "^1.0.14",
-    "llamaindex": "^0.9.1",
+    "llamaindex": "^0.9.8",
    "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",
@@ -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));
@@ -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));
@@ -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";

@@ -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({
@@ -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({
@@ -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",
@@ -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);
@@ -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",
@@ -1,5 +1,44 @@
 # @llamaindex/autotool

+## 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
+
+- llamaindex@0.9.3
+
 ## 6.0.2

 ### Patch Changes
@@ -1,5 +1,50 @@
 # @llamaindex/autotool-01-node-example

+## 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
+
+- llamaindex@0.9.3
+- @llamaindex/autotool@6.0.3
+
 ## 0.0.83

 ### Patch Changes
@@ -13,5 +13,5 @@
  "scripts": {
    "start": "node --import tsx --import @llamaindex/autotool/node ./src/index.ts"
  },
-  "version": "0.0.83"
+  "version": "0.0.89"
 }
@@ -1,5 +1,50 @@
 # @llamaindex/autotool-02-next-example

+## 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
+
+- llamaindex@0.9.3
+- @llamaindex/autotool@6.0.3
+
 ## 0.1.127

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/autotool-02-next-example",
  "private": true,
-  "version": "0.1.127",
+  "version": "0.1.133",
  "scripts": {
    "dev": "next dev",
    "build": "next build",
@@ -6,7 +6,7 @@
    "url": "git+https://github.com/run-llama/LlamaIndexTS.git",
    "directory": "packages/autotool"
  },
-  "version": "6.0.2",
+  "version": "6.0.8",
  "description": "auto transpile your JS function to LLM Agent compatible",
  "files": [
    "dist",
@@ -1,5 +1,41 @@
 # @llamaindex/cloud

+## 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
+
+- Updated dependencies [d952e68]
+  - @llamaindex/core@0.5.2
+
 ## 3.0.2

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/cloud",
-  "version": "3.0.2",
+  "version": "3.0.7",
  "type": "module",
  "license": "MIT",
  "scripts": {
@@ -1,5 +1,43 @@
 # @llamaindex/community

+## 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
+
+- Updated dependencies [d952e68]
+  - @llamaindex/core@0.5.2
+
 ## 0.0.83

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/community",
  "description": "Community package for LlamaIndexTS",
-  "version": "0.0.83",
+  "version": "0.0.88",
  "type": "module",
  "types": "dist/type/index.d.ts",
  "main": "dist/cjs/index.js",
@@ -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;
@@ -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 = {
@@ -1,5 +1,37 @@
 # @llamaindex/core

+## 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
+
+- d952e68: Refine synthesizer will now return an empty string as the response if an empty array of source nodes were provided. Before it would throw an internal error converting undefined to ReadableStream.
+
 ## 0.5.1

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/core",
  "type": "module",
-  "version": "0.5.1",
+  "version": "0.5.6",
  "description": "LlamaIndex Core Module",
  "exports": {
    "./agent": {
@@ -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();
@@ -215,6 +215,10 @@ 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;
 };

 /**
@@ -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[] {
@@ -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);
@@ -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";
@@ -116,7 +116,14 @@ class Refine extends BaseSynthesizer {
      }
    }

-    // fixme: no source nodes provided, cannot fix right now due to lack of context
+    if (response === undefined) {
+      response = stream
+        ? (async function* () {
+            yield "";
+          })()
+        : "";
+    }
+
    if (typeof response === "string") {
      return EngineResponse.fromResponse(response, false, nodes);
    } else {
@@ -59,14 +59,28 @@ export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
  }

  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 });
+        } else {
+          return this.#fn.call(null, result.data);
+        }
      } else {
        console.warn(result.error.errors);
      }
    }
+
    return this.#fn.call(null, input);
  };
 }
@@ -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> => {
@@ -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);
+});
@@ -66,9 +66,9 @@ describe("ChatMemoryBuffer", () => {
    expect(result).toEqual([...inputMessages, ...storedMessages]);
  });

-  test("getMessages throws error when initial token count exceeds limit", () => {
+  test("getMessages throws error when initial token count exceeds limit", async () => {
    const buffer = new ChatMemoryBuffer({ tokenLimit: 10 });
-    expect(async () => buffer.getMessages(undefined, 20)).rejects.toThrow(
+    await expect(async () => buffer.getMessages(undefined, 20)).rejects.toThrow(
      "Initial token count exceeds token limit",
    );
  });
@@ -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);
+  });
+});
@@ -1,4 +1,4 @@
-import { describe, expect, test, vi } from "vitest";
+import { beforeEach, describe, expect, test, vi } from "vitest";
 import type { LLMMetadata } from "../../llms/dist/index.js";
 import { getResponseSynthesizer } from "../../response-synthesizers/dist/index.js";
 import { Document } from "../../schema/dist/index.js";
@@ -10,26 +10,69 @@ const mockLllm = () => ({
      return response;
    }

-    function* gen() {
-      // yield a few times to make sure each chunk has the sourceNodes
-      yield response;
-      yield response;
-      yield response;
-    }
-
-    return gen();
+    return {
+      [Symbol.asyncIterator]: function* gen() {
+        // yield a few times to make sure each chunk has the sourceNodes
+        yield response;
+        yield response;
+        yield response;
+      },
+    };
  }),
  chat: vi.fn(),
  metadata: {} as unknown as LLMMetadata,
 });

+describe("refine response synthesizer", () => {
+  let synthesizer: ReturnType<typeof getResponseSynthesizer<"refine">>;
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  const isAsyncIterable = (obj: any): boolean =>
+    obj[Symbol.asyncIterator] !== undefined;
+
+  beforeEach(() => {
+    synthesizer = getResponseSynthesizer("refine", {
+      llm: mockLllm(),
+    });
+  });
+
+  describe("getResponse", () => {
+    test("should return async iterable of EngineResponse when stream is true and sourceNodes are empty", async () => {
+      const response = await synthesizer.getResponse(
+        "unimportant query",
+        [],
+        true,
+      );
+
+      expect(isAsyncIterable(response)).toBe(true);
+      for await (const chunk of response) {
+        expect(chunk.message.content).toEqual("");
+      }
+    });
+
+    test("should return non async iterable when stream is false and sourceNodes are empty", async () => {
+      const response = await synthesizer.getResponse(
+        "unimportant query",
+        [],
+        false,
+      );
+
+      expect(isAsyncIterable(response)).toBe(false);
+      expect(response.message.content).toEqual("");
+    });
+  });
+});
+
 describe("compact and refine response synthesizer", () => {
+  let synthesizer: ReturnType<typeof getResponseSynthesizer<"compact">>;
+
+  beforeEach(() => {
+    synthesizer = getResponseSynthesizer("compact", {
+      llm: mockLllm(),
+    });
+  });
+
  describe("synthesize", () => {
    test("should return original sourceNodes with response when stream = false", async () => {
-      const synthesizer = getResponseSynthesizer("compact", {
-        llm: mockLllm(),
-      });
-
      const sourceNode = { node: new Document({}), score: 1 };

      const response = await synthesizer.synthesize(
@@ -44,10 +87,6 @@ describe("compact and refine response synthesizer", () => {
    });

    test("should return original sourceNodes with response when stream = true", async () => {
-      const synthesizer = getResponseSynthesizer("compact", {
-        llm: mockLllm(),
-      });
-
      const sourceNode = { node: new Document({}), score: 1 };

      const response = await synthesizer.synthesize(
@@ -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
@@ -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
    }
  }
@@ -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 {
@@ -1,5 +1,44 @@
 # @llamaindex/experimental

+## 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
+
+- llamaindex@0.9.3
+
 ## 0.0.152

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/experimental",
  "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.152",
+  "version": "0.0.158",
  "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",
@@ -1,5 +1,69 @@
 # llamaindex

+## 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
+
+- Updated dependencies [d952e68]
+  - @llamaindex/core@0.5.2
+  - @llamaindex/cloud@3.0.3
+  - @llamaindex/node-parser@1.0.2
+  - @llamaindex/openai@0.1.54
+
 ## 0.9.2

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "llamaindex",
-  "version": "0.9.2",
+  "version": "0.9.8",
  "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",
@@ -1 +0,0 @@
-export * from "@llamaindex/openai";
@@ -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";
@@ -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";
@@ -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";
@@ -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,
@@ -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,
@@ -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,
@@ -1,4 +0,0 @@
-export { DeepSeekLLM } from "./deepseek.js";
-export { FireworksLLM } from "./fireworks.js";
-export * from "./openai.js";
-export { TogetherLLM } from "./together.js";
@@ -1 +0,0 @@
-export * from "@llamaindex/openai";
@@ -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 {
@@ -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";

@@ -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(() => {
@@ -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();
+  });
+});
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
github-actions[bot]	a3cbcb31a2	Release 0.9.8 (#1711 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: marcusschiesser <17126+marcusschiesser@users.noreply.github.com>	2025-03-06 16:33:19 +07:00
Thuc Pham	bbc8c8787d	fix: prefer using embedding model from vector store (#1708 )	2025-03-06 16:24:05 +07:00
Huu Le	4b49428f57	fix agent workflow tool call for Ollama (#1706 ) Co-authored-by: Marcus Schiesser <mail@marcusschiesser.de>	2025-03-06 11:13:42 +07:00
Peter Goldstein	7ee4968b06	Add Gemini 2.0 Pro Experimental (#1707 )	2025-03-06 11:04:56 +07:00
github-actions[bot]	0111f5c8b0	Release 0.9.7 (#1703 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: marcusschiesser <17126+marcusschiesser@users.noreply.github.com>	2025-03-05 16:59:05 +07:00
Thuc Pham	beb922b743	fix: build fail in edge runtime (#1705 ) Co-authored-by: Marcus Schiesser <mail@marcusschiesser.de>	2025-03-05 16:35:00 +07:00
patryktop	e28c29d1f5	feat: add Llama 3.3 70B Instruct to community package (#1702 )	2025-03-04 17:27:35 -08:00
github-actions[bot]	008cccd9f1	Release 0.9.6 (#1698 )	2025-03-04 17:33:18 +07:00
Huu Le	081698d68c	chore: simplify imports of agent workflow (#1700 )	2025-03-04 17:01:29 +07:00
Huu Le	ab5fe5d7a0	chore: remove core import in document (#1699 )	2025-03-04 16:14:31 +07:00
Huu Le	56689707d3	feat: Support AgentWorkflow (#1685 ) Co-authored-by: Marcus Schiesser <mail@marcusschiesser.de>	2025-03-04 16:05:25 +07:00
Brian Lange	fd74ba4bf1	fix: Voyage typescript configs + docs (#1696 )	2025-03-04 11:00:05 +07:00
github-actions[bot]	b2634e47ca	Release 0.9.5 (#1694 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: marcusschiesser <17126+marcusschiesser@users.noreply.github.com>	2025-02-28 18:18:14 +07:00
Thuc Pham	ad3c7f1ec1	fix: streaming issues with LLMAgent (#1692 )	2025-02-28 18:13:36 +07:00
Alex Yang	335f2df626	chore: fix lock file	2025-02-27 12:31:11 -08:00
github-actions[bot]	ee963644bf	Release 0.9.4 (#1689 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>	2025-02-27 12:28:48 -08:00
Alex Yang	cb256f24ae	feat: support gpt-4.5 (#1688 )	2025-02-27 12:24:33 -08:00
Alex Yang	1ccc04ecb5	chore: fix changeset	2025-02-27 12:17:12 -08:00
Brian Lange	034639153b	feat: Voyage embeddings (#1574 ) Co-authored-by: Marcus Schiesser <mail@marcusschiesser.de>	2025-02-27 10:22:33 +07:00
patryktop	1914b52708	feat: add Claude 3.7 Sonnet model to community package (#1683 )	2025-02-26 14:38:14 -08:00
Alex Yang	cb021e7196	feat(node-parser): support async function (#1682 )	2025-02-26 08:59:51 -08:00
ratacat	c2aa836b35	docs: upgrade remote ollama embeddings (#1680 )	2025-02-25 11:39:16 -08:00
github-actions[bot]	3b0f55f1ea	Release 0.9.3 (#1676 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: marcusschiesser <17126+marcusschiesser@users.noreply.github.com>	2025-02-25 16:53:30 +07:00
Marcus Schiesser	a9c6144eec	feat: stream thinking tokens for claude 3.7 (#1679 )	2025-02-25 16:48:12 +07:00
Marcus Schiesser	3564244ced	chore: add claude 3.7 support (#1678 )	2025-02-25 15:18:45 +07:00
Gunnar Holwerda	d952e68ec4	Fix refine synthesizer empty source nodes behavior (#1677 )	2025-02-24 10:56:21 +07:00
Tushar Sonawane	5c026e839f	feat(vectorstore): adds firestore vector store support (#1600 )	2025-02-21 16:23:50 +07:00