Release 0.8.13 (#1480 )

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: himself65 <himself65@users.noreply.github.com>
docs: update docusaurus.config.js (#1489 )
2026-07-01 22:14:03 -04:00 · 2024-11-14 18:25:19 -08:00 · 2024-11-14 17:36:52 -08:00 · 2024-11-14 17:35:21 -08:00 · 2024-11-14 15:46:48 -08:00 · 2024-11-14 13:45:50 -08:00
100 changed files with 2061 additions and 337 deletions
@@ -1,12 +1,16 @@
-# LlamaIndex.TS
+<p align="center">
+  <img height="100" width="100" alt="LlamaIndex logo" src="https://ts.llamaindex.ai/square.svg" />
+</p>
+<h1 align="center">LlamaIndex.TS</h1>
+<h3 align="center">
+  Data framework for your LLM application.
+</h3>

 [![NPM Version](https://img.shields.io/npm/v/llamaindex)](https://www.npmjs.com/package/llamaindex)
 [![NPM License](https://img.shields.io/npm/l/llamaindex)](https://www.npmjs.com/package/llamaindex)
 [![NPM Downloads](https://img.shields.io/npm/dm/llamaindex)](https://www.npmjs.com/package/llamaindex)
 [![Discord](https://img.shields.io/discord/1059199217496772688)](https://discord.com/invite/eN6D2HQ4aX)

-LlamaIndex is a data framework for your LLM application.
-
 Use your own data with large language models (LLMs, OpenAI ChatGPT and others) in JS runtime environments with TypeScript support.

 Documentation: https://ts.llamaindex.ai/
@@ -1,5 +1,38 @@
 # docs

+## 0.0.118
+
+### Patch Changes
+
+- llamaindex@0.8.13
+- @llamaindex/examples@0.0.16
+
+## 0.0.117
+
+### Patch Changes
+
+- @llamaindex/examples@0.0.15
+
+## 0.0.116
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.0.115
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.0.114
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+  - @llamaindex/examples@0.0.14
+
 ## 0.0.113

 ### Patch Changes
@@ -62,6 +62,12 @@ const config = {
    ({
      // Replace with your project's social card
      image: "img/favicon.png", // TODO change this
+      announcementBar: {
+        id: "migrate_to_next",
+        content:
+          'We are migrating to Next.js based documentation. Check it out <a href="https://ts.llamaindex.ai/docs/llamaindex">here</a>!',
+        isCloseable: false,
+      },
      navbar: {
        title: "LlamaIndex.TS",
        logo: {
@@ -1,6 +1,6 @@
 {
  "name": "docs",
-  "version": "0.0.113",
+  "version": "0.0.118",
  "private": true,
  "scripts": {
    "docusaurus": "docusaurus",
@@ -1,5 +1,68 @@
 # @llamaindex/doc

+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [ee20c44]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+  - @llamaindex/workflow@0.0.6
+  - llamaindex@0.8.13
+  - @llamaindex/cloud@2.0.10
+  - @llamaindex/node-parser@0.0.11
+  - @llamaindex/openai@0.1.35
+  - @llamaindex/readers@1.0.11
+
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [ea92b69]
+- Updated dependencies [fadc8b8]
+  - @llamaindex/workflow@0.0.5
+
+## 0.0.14
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+  - @llamaindex/openai@0.1.34
+  - @llamaindex/cloud@2.0.9
+  - llamaindex@0.8.12
+  - @llamaindex/node-parser@0.0.10
+  - @llamaindex/readers@1.0.10
+
+## 0.0.13
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+  - @llamaindex/cloud@2.0.8
+  - llamaindex@0.8.11
+  - @llamaindex/node-parser@0.0.9
+  - @llamaindex/openai@0.1.33
+  - @llamaindex/readers@1.0.9
+
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+- Updated dependencies [fd8c882]
+  - llamaindex@0.8.10
+  - @llamaindex/core@0.4.7
+  - @llamaindex/workflow@0.0.4
+  - @llamaindex/cloud@2.0.7
+  - @llamaindex/node-parser@0.0.8
+  - @llamaindex/openai@0.1.32
+  - @llamaindex/readers@1.0.8
+
 ## 0.0.11

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/doc",
-  "version": "0.0.11",
+  "version": "0.0.16",
  "private": true,
  "scripts": {
    "build": "pnpm run build:docs && next build",
@@ -1,6 +1,14 @@
 {
  "extends": ["//"],
  "tasks": {
+    "build": {
+      "outputs": [
+        ".next",
+        ".source",
+        "next-env.d.ts",
+        "src/content/docs/cloud/api/**"
+      ]
+    },
    "dev": {
      "dependsOn": ["^build"]
    }
@@ -1,5 +1,30 @@
 # @llamaindex/cloudflare-worker-agent-test

+## 0.0.109
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 0.0.108
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.0.107
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.0.106
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 0.0.105

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.105",
+  "version": "0.0.109",
  "type": "module",
  "private": true,
  "scripts": {
@@ -1,5 +1,29 @@
 # @llamaindex/llama-parse-browser-test

+## 0.0.30
+
+### Patch Changes
+
+- @llamaindex/cloud@2.0.10
+
+## 0.0.29
+
+### Patch Changes
+
+- @llamaindex/cloud@2.0.9
+
+## 0.0.28
+
+### Patch Changes
+
+- @llamaindex/cloud@2.0.8
+
+## 0.0.27
+
+### Patch Changes
+
+- @llamaindex/cloud@2.0.7
+
 ## 0.0.26

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/llama-parse-browser-test",
  "private": true,
-  "version": "0.0.26",
+  "version": "0.0.30",
  "type": "module",
  "scripts": {
    "dev": "vite",
@@ -1,5 +1,30 @@
 # @llamaindex/next-agent-test

+## 0.1.109
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 0.1.108
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.1.107
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.1.106
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 0.1.105

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/next-agent-test",
-  "version": "0.1.105",
+  "version": "0.1.109",
  "private": true,
  "scripts": {
    "dev": "next dev",
@@ -1,5 +1,30 @@
 # test-edge-runtime

+## 0.1.108
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 0.1.107
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.1.106
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.1.105
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 0.1.104

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.104",
+  "version": "0.1.108",
  "private": true,
  "scripts": {
    "dev": "next dev",
@@ -1,5 +1,30 @@
 # @llamaindex/next-node-runtime

+## 0.0.90
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 0.0.89
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.0.88
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.0.87
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 0.0.86

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.0.86",
+  "version": "0.0.90",
  "private": true,
  "scripts": {
    "dev": "next dev",
@@ -1,5 +1,30 @@
 # @llamaindex/waku-query-engine-test

+## 0.0.109
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 0.0.108
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.0.107
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.0.106
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 0.0.105

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.105",
+  "version": "0.0.109",
  "type": "module",
  "private": true,
  "scripts": {
@@ -167,6 +167,7 @@ For questions about more specific sections, please use the vector_tool.`,
  const mockCall = t.mock.fn(({ query }: { query: string }) => {
    return originalCall({ query });
  });
+  // @ts-expect-error what?
  queryEngineTools[1]!.call = mockCall;

  const toolMapping = SimpleToolNodeMapping.fromObjects(queryEngineTools);
@@ -1,5 +1,38 @@
 # examples

+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [ee20c44]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+  - @llamaindex/workflow@0.0.6
+  - llamaindex@0.8.13
+  - @llamaindex/readers@1.0.11
+
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [ea92b69]
+- Updated dependencies [fadc8b8]
+  - @llamaindex/workflow@0.0.5
+
+## 0.0.14
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+- Updated dependencies [fd8c882]
+  - llamaindex@0.8.10
+  - @llamaindex/core@0.4.7
+  - @llamaindex/workflow@0.0.4
+  - @llamaindex/readers@1.0.8
+
 ## 0.0.13

 ### Patch Changes
@@ -0,0 +1,38 @@
+import { Anthropic } from "llamaindex";
+
+async function main() {
+  const anthropic = new Anthropic({
+    model: "claude-3-5-sonnet-20241022",
+  });
+
+  const entireBook = await fetch(
+    "https://www.gutenberg.org/files/1342/1342-0.txt",
+  ).then((response) => response.text());
+
+  const response = await anthropic.chat({
+    messages: [
+      {
+        content:
+          "You are an AI assistant tasked with analyzing literary works. Your goal is to provide insightful commentary on themes, characters, and writing style.\n",
+        role: "system",
+      },
+      {
+        content: entireBook,
+        role: "system",
+        options: {
+          cache_control: {
+            type: "ephemeral",
+          },
+        },
+      },
+      {
+        content: "analyze the major themes in Pride and Prejudice.",
+        role: "user",
+      },
+    ],
+  });
+
+  console.log(response.message.content);
+}
+
+main().catch(console.error);
@@ -1,15 +1,15 @@
 {
  "name": "@llamaindex/examples",
  "private": true,
-  "version": "0.0.13",
+  "version": "0.0.16",
  "dependencies": {
    "@aws-crypto/sha256-js": "^5.2.0",
    "@azure/cosmos": "^4.1.1",
    "@azure/identity": "^4.4.1",
    "@datastax/astra-db-ts": "^1.4.1",
-    "@llamaindex/core": "^0.4.5",
-    "@llamaindex/readers": "^1.0.6",
-    "@llamaindex/workflow": "^0.0.3",
+    "@llamaindex/core": "^0.4.10",
+    "@llamaindex/readers": "^1.0.11",
+    "@llamaindex/workflow": "^0.0.6",
    "@notionhq/client": "^2.2.15",
    "@pinecone-database/pinecone": "^3.0.2",
    "@vercel/postgres": "^0.10.0",
@@ -18,7 +18,7 @@
    "commander": "^12.1.0",
    "dotenv": "^16.4.5",
    "js-tiktoken": "^1.0.14",
-    "llamaindex": "^0.8.8",
+    "llamaindex": "^0.8.13",
    "mongodb": "^6.7.0",
    "pathe": "^1.1.2",
    "postgres": "^3.4.4"
@@ -0,0 +1,16 @@
+import { VLLM } from "llamaindex";
+
+const llm = new VLLM({
+  model: "NousResearch/Meta-Llama-3-8B-Instruct",
+});
+
+const response = await llm.chat({
+  messages: [
+    {
+      role: "user",
+      content: "Hello?",
+    },
+  ],
+});
+
+console.log(response.message.content);
@@ -1,14 +1,19 @@
 import {
-  Context,
+  HandlerContext,
  StartEvent,
  StopEvent,
  Workflow,
  WorkflowEvent,
-} from "@llamaindex/core/workflow";
+} from "@llamaindex/workflow";
 import { OpenAI } from "llamaindex";

 const MAX_REVIEWS = 3;

+type Context = {
+  specification: string;
+  numberReviews: number;
+};
+
 // Using the o1-preview model (see https://platform.openai.com/docs/guides/reasoning?reasoning-prompt-examples=coding-planning)
 const llm = new OpenAI({ model: "o1-preview", temperature: 1 });

@@ -20,7 +25,9 @@ stores the question/answer pair in the database.`;

 // Create custom event types
 export class MessageEvent extends WorkflowEvent<{ msg: string }> {}
+
 export class CodeEvent extends WorkflowEvent<{ code: string }> {}
+
 export class ReviewEvent extends WorkflowEvent<{
  review: string;
  code: string;
@@ -34,12 +41,13 @@ const truncate = (str: string) => {
 };

 // the architect is responsible for writing the structure and the initial code based on the specification
-const architect = async (context: Context, ev: StartEvent) => {
-  // get the specification from the start event and save it to context
-  context.set("specification", ev.data.input);
-  const spec = context.get("specification");
+const architect = async (
+  context: HandlerContext<Context>,
+  _: StartEvent<string>,
+) => {
+  const spec = context.data.specification;
  // write a message to send an update to the user
-  context.writeEventToStream(
+  context.sendEvent(
    new MessageEvent({
      msg: `Writing app using this specification: ${truncate(spec)}`,
    }),
@@ -50,13 +58,13 @@ const architect = async (context: Context, ev: StartEvent) => {
 };

 // the coder is responsible for updating the code based on the review
-const coder = async (context: Context, ev: ReviewEvent) => {
+const coder = async (context: HandlerContext<Context>, ev: ReviewEvent) => {
  // get the specification from the context
-  const spec = context.get("specification");
+  const spec = context.data.specification;
  // get the latest review and code
  const { review, code } = ev.data;
  // write a message to send an update to the user
-  context.writeEventToStream(
+  context.sendEvent(
    new MessageEvent({
      msg: `Update code based on review: ${truncate(review)}`,
    }),
@@ -67,32 +75,35 @@ const coder = async (context: Context, ev: ReviewEvent) => {
 };

 // the reviewer is responsible for reviewing the code and providing feedback
-const reviewer = async (context: Context, ev: CodeEvent) => {
+const reviewer = async (context: HandlerContext<Context>, ev: CodeEvent) => {
  // get the specification from the context
-  const spec = context.get("specification");
+  const spec = context.data.specification;
  // get latest code from the event
  const { code } = ev.data;
  // update and check the number of reviews
-  const numberReviews = context.get("numberReviews", 0) + 1;
-  context.set("numberReviews", numberReviews);
-  if (numberReviews > MAX_REVIEWS) {
+  context.data.numberReviews++;
+  if (context.data.numberReviews > MAX_REVIEWS) {
    // the we've done this too many times - return the code
-    context.writeEventToStream(
+    context.sendEvent(
      new MessageEvent({
-        msg: `Already reviewed ${numberReviews - 1} times, stopping!`,
+        msg: `Already reviewed ${
+          context.data.numberReviews - 1
+        } times, stopping!`,
      }),
    );
    return new StopEvent({ result: code });
  }
  // write a message to send an update to the user
-  context.writeEventToStream(
-    new MessageEvent({ msg: `Review #${numberReviews}: ${truncate(code)}` }),
+  context.sendEvent(
+    new MessageEvent({
+      msg: `Review #${context.data.numberReviews}: ${truncate(code)}`,
+    }),
  );
  const prompt = `Review this code: <code>${code}</code>. Check if the code quality and whether it correctly implements this specification: <spec>${spec}</spec>. If you're satisfied, just return 'Looks great', nothing else. If not, return a review with a list of changes you'd like to see.`;
  const review = (await llm.complete({ prompt })).text;
  if (review.includes("Looks great")) {
    // the reviewer is satisfied with the code, let's return the review
-    context.writeEventToStream(
+    context.sendEvent(
      new MessageEvent({
        msg: `Reviewer says: ${review}`,
      }),
@@ -103,20 +114,44 @@ const reviewer = async (context: Context, ev: CodeEvent) => {
  return new ReviewEvent({ review, code });
 };

-const codeAgent = new Workflow({ validate: true });
-codeAgent.addStep(StartEvent, architect, { outputs: CodeEvent });
-codeAgent.addStep(ReviewEvent, coder, { outputs: CodeEvent });
-codeAgent.addStep(CodeEvent, reviewer, { outputs: ReviewEvent });
+const codeAgent = new Workflow<Context, string, string>();
+codeAgent.addStep(
+  {
+    inputs: [StartEvent<string>],
+    outputs: [CodeEvent],
+  },
+  architect,
+);
+codeAgent.addStep(
+  {
+    inputs: [ReviewEvent],
+    outputs: [CodeEvent],
+  },
+  coder,
+);
+codeAgent.addStep(
+  {
+    inputs: [CodeEvent],
+    outputs: [ReviewEvent, StopEvent],
+  },
+  reviewer,
+);

 // Usage
 async function main() {
-  const run = codeAgent.run(specification);
-  for await (const event of codeAgent.streamEvents()) {
-    const msg = (event as MessageEvent).data.msg;
-    console.log(`${msg}\n`);
+  const run = codeAgent.run(specification).with({
+    specification,
+    numberReviews: 0,
+  });
+  for await (const event of run) {
+    if (event instanceof MessageEvent) {
+      const msg = (event as MessageEvent).data.msg;
+      console.log(`${msg}\n`);
+    } else if (event instanceof StopEvent) {
+      const result = (event as StopEvent<string>).data;
+      console.log("Final code:\n", result);
+    }
  }
-  const result = await run;
-  console.log("Final code:\n", result.data.result);
 }

 main().catch(console.error);
@@ -1,10 +1,10 @@
 import {
-  Context,
+  HandlerContext,
  StartEvent,
  StopEvent,
  Workflow,
  WorkflowEvent,
-} from "@llamaindex/core/workflow";
+} from "@llamaindex/workflow";
 import { OpenAI } from "llamaindex";

 // Create LLM instance
@@ -12,59 +12,77 @@ const llm = new OpenAI();

 // Create custom event types
 export class JokeEvent extends WorkflowEvent<{ joke: string }> {}
+
 export class CritiqueEvent extends WorkflowEvent<{ critique: string }> {}
+
 export class AnalysisEvent extends WorkflowEvent<{ analysis: string }> {}

-const generateJoke = async (_context: Context, ev: StartEvent) => {
-  const prompt = `Write your best joke about ${ev.data.input}.`;
+const generateJoke = async (_: unknown, ev: StartEvent<string>) => {
+  const prompt = `Write your best joke about ${ev.data}.`;
  const response = await llm.complete({ prompt });
  return new JokeEvent({ joke: response.text });
 };

-const critiqueJoke = async (_context: Context, ev: JokeEvent) => {
+const critiqueJoke = async (_: unknown, ev: JokeEvent) => {
  const prompt = `Give a thorough critique of the following joke: ${ev.data.joke}`;
  const response = await llm.complete({ prompt });
  return new CritiqueEvent({ critique: response.text });
 };

-const analyzeJoke = async (_context: Context, ev: JokeEvent) => {
+const analyzeJoke = async (_: unknown, ev: JokeEvent) => {
  const prompt = `Give a thorough analysis of the following joke: ${ev.data.joke}`;
  const response = await llm.complete({ prompt });
  return new AnalysisEvent({ analysis: response.text });
 };

 const reportJoke = async (
-  context: Context,
-  ev: AnalysisEvent | CritiqueEvent,
+  context: HandlerContext,
+  ev1: AnalysisEvent,
+  ev2: CritiqueEvent,
 ) => {
-  const events = context.collectEvents(ev, [AnalysisEvent, CritiqueEvent]);
-  if (!events) {
-    return;
-  }
-  const subPrompts = events.map((event) => {
-    if (event instanceof AnalysisEvent) {
-      return `Analysis: ${event.data.analysis}`;
-    } else if (event instanceof CritiqueEvent) {
-      return `Critique: ${event.data.critique}`;
-    }
-    return "";
-  });
+  const subPrompts = [ev1.data.analysis, ev2.data.critique];

-  const prompt = `Based on the following information about a joke:\n${subPrompts.join("\n")}\nProvide a comprehensive report on the joke's quality and impact.`;
+  const prompt = `Based on the following information about a joke:\n${subPrompts.join(
+    "\n",
+  )}\nProvide a comprehensive report on the joke's quality and impact.`;
  const response = await llm.complete({ prompt });
-  return new StopEvent({ result: response.text });
+  return new StopEvent(response.text);
 };

-const jokeFlow = new Workflow();
-jokeFlow.addStep(StartEvent, generateJoke);
-jokeFlow.addStep(JokeEvent, critiqueJoke);
-jokeFlow.addStep(JokeEvent, analyzeJoke);
-jokeFlow.addStep([AnalysisEvent, CritiqueEvent], reportJoke);
+const jokeFlow = new Workflow<unknown, string, string>();
+jokeFlow.addStep(
+  {
+    inputs: [StartEvent<string>],
+    outputs: [JokeEvent],
+  },
+  generateJoke,
+);
+jokeFlow.addStep(
+  {
+    inputs: [JokeEvent],
+    outputs: [CritiqueEvent],
+  },
+  critiqueJoke,
+);
+jokeFlow.addStep(
+  {
+    inputs: [JokeEvent],
+    outputs: [AnalysisEvent],
+  },
+  analyzeJoke,
+);
+jokeFlow.addStep(
+  {
+    inputs: [AnalysisEvent, CritiqueEvent],
+    outputs: [StopEvent<string>],
+  },
+  reportJoke,
+);

 // Usage
 async function main() {
  const result = await jokeFlow.run("pirates");
-  console.log(result.data.result);
+  console.log(result.data);
 }

 main().catch(console.error);
@@ -1,10 +1,9 @@
 import {
-  Context,
  StartEvent,
  StopEvent,
  Workflow,
  WorkflowEvent,
-} from "@llamaindex/core/workflow";
+} from "@llamaindex/workflow";
 import { OpenAI } from "llamaindex";

 // Create LLM instance
@@ -13,26 +12,38 @@ const llm = new OpenAI();
 // Create a custom event type
 export class JokeEvent extends WorkflowEvent<{ joke: string }> {}

-const generateJoke = async (_context: Context, ev: StartEvent) => {
-  const prompt = `Write your best joke about ${ev.data.input}.`;
+const generateJoke = async (_: unknown, ev: StartEvent<string>) => {
+  const prompt = `Write your best joke about ${ev.data}.`;
  const response = await llm.complete({ prompt });
  return new JokeEvent({ joke: response.text });
 };

-const critiqueJoke = async (_context: Context, ev: JokeEvent) => {
+const critiqueJoke = async (_: unknown, ev: JokeEvent) => {
  const prompt = `Give a thorough critique of the following joke: ${ev.data.joke}`;
  const response = await llm.complete({ prompt });
-  return new StopEvent({ result: response.text });
+  return new StopEvent(response.text);
 };

-const jokeFlow = new Workflow({ verbose: true });
-jokeFlow.addStep(StartEvent, generateJoke);
-jokeFlow.addStep(JokeEvent, critiqueJoke);
+const jokeFlow = new Workflow<unknown, string, string>();
+jokeFlow.addStep(
+  {
+    inputs: [StartEvent<string>],
+    outputs: [JokeEvent],
+  },
+  generateJoke,
+);
+jokeFlow.addStep(
+  {
+    inputs: [JokeEvent],
+    outputs: [StopEvent<string>],
+  },
+  critiqueJoke,
+);

 // Usage
 async function main() {
  const result = await jokeFlow.run("pirates");
-  console.log(result.data.result);
+  console.log(result.data);
 }

 main().catch(console.error);
@@ -1,10 +1,10 @@
 import {
-  Context,
+  HandlerContext,
  StartEvent,
  StopEvent,
  Workflow,
  WorkflowEvent,
-} from "@llamaindex/core/workflow";
+} from "@llamaindex/workflow";
 import { OpenAI } from "llamaindex";

 // Create LLM instance
@@ -12,38 +12,55 @@ const llm = new OpenAI();

 // Create custom event types
 export class JokeEvent extends WorkflowEvent<{ joke: string }> {}
+
 export class MessageEvent extends WorkflowEvent<{ msg: string }> {}

-const generateJoke = async (context: Context, ev: StartEvent) => {
-  context.writeEventToStream(
-    new MessageEvent({ msg: `Generating a joke about: ${ev.data.input}` }),
+const generateJoke = async (context: HandlerContext, ev: StartEvent) => {
+  context.sendEvent(
+    new MessageEvent({ msg: `Generating a joke about: ${ev.data}` }),
  );
-  const prompt = `Write your best joke about ${ev.data.input}.`;
+  const prompt = `Write your best joke about ${ev.data}.`;
  const response = await llm.complete({ prompt });
  return new JokeEvent({ joke: response.text });
 };

-const critiqueJoke = async (context: Context, ev: JokeEvent) => {
-  context.writeEventToStream(
+const critiqueJoke = async (context: HandlerContext, ev: JokeEvent) => {
+  context.sendEvent(
    new MessageEvent({ msg: `Write a critique of this joke: ${ev.data.joke}` }),
  );
  const prompt = `Give a thorough critique of the following joke: ${ev.data.joke}`;
  const response = await llm.complete({ prompt });
-  return new StopEvent({ result: response.text });
+  return new StopEvent(response.text);
 };

 const jokeFlow = new Workflow();
-jokeFlow.addStep(StartEvent, generateJoke);
-jokeFlow.addStep(JokeEvent, critiqueJoke);
+jokeFlow.addStep(
+  {
+    inputs: [StartEvent<string>],
+    outputs: [JokeEvent],
+  },
+  generateJoke,
+);
+jokeFlow.addStep(
+  {
+    inputs: [JokeEvent],
+    outputs: [StopEvent<string>],
+  },
+  critiqueJoke,
+);

 // Usage
 async function main() {
  const run = jokeFlow.run("pirates");
-  for await (const event of jokeFlow.streamEvents()) {
-    console.log((event as MessageEvent).data.msg);
+  for await (const event of run) {
+    if (event instanceof MessageEvent) {
+      console.log("Message:");
+      console.log((event as MessageEvent).data.msg);
+    } else if (event instanceof StopEvent) {
+      console.log("Result:");
+      console.log((event as StopEvent<string>).data);
+    }
  }
-  const result = await run;
-  console.log(result.data.result);
 }

 main().catch(console.error);
@@ -1,19 +1,21 @@
-import {
-  Context,
-  StartEvent,
-  StopEvent,
-  Workflow,
-} from "@llamaindex/core/workflow";
+import { StartEvent, StopEvent, Workflow } from "@llamaindex/workflow";

-const longRunning = async (_context: Context, ev: StartEvent) => {
+const longRunning = async (_: unknown, ev: StartEvent<string>) => {
  await new Promise((resolve) => setTimeout(resolve, 2000)); // Wait for 2 seconds
-  return new StopEvent({ result: "We waited 2 seconds" });
+  return new StopEvent("We waited 2 seconds");
 };

 async function timeout() {
-  const workflow = new Workflow({ verbose: true, timeout: 1 });
-  workflow.addStep(StartEvent, longRunning);
-  // This will timeout
+  const workflow = new Workflow<unknown, string, string>({
+    timeout: 1,
+  });
+  workflow.addStep(
+    {
+      inputs: [StartEvent<string>],
+      outputs: [StopEvent<string>],
+    },
+    longRunning,
+  );
  try {
    await workflow.run("Let's start");
  } catch (error) {
@@ -23,14 +25,23 @@ async function timeout() {

 async function notimeout() {
  // Increase timeout to 3 seconds - no timeout
-  const workflow = new Workflow({ verbose: true, timeout: 3 });
-  workflow.addStep(StartEvent, longRunning);
+  const workflow = new Workflow<unknown, string, string>({
+    timeout: 3,
+  });
+  workflow.addStep(
+    {
+      inputs: [StartEvent<string>],
+      outputs: [StopEvent<string>],
+    },
+    longRunning,
+  );
  const result = await workflow.run("Let's start");
-  console.log(result.data.result);
+  console.log(result.data);
 }

 async function main() {
  await timeout();
+  console.log("---");
  await notimeout();
 }

@@ -1,10 +1,9 @@
 import {
-  Context,
  StartEvent,
  StopEvent,
  Workflow,
  WorkflowEvent,
-} from "@llamaindex/core/workflow";
+} from "@llamaindex/workflow";
 import { OpenAI } from "llamaindex";

 // Create LLM instance
@@ -13,40 +12,66 @@ const llm = new OpenAI();
 // Create a custom event type
 export class JokeEvent extends WorkflowEvent<{ joke: string }> {}

-const generateJoke = async (_context: Context, ev: StartEvent) => {
-  const prompt = `Write your best joke about ${ev.data.input}.`;
+const generateJoke = async (_: unknown, ev: StartEvent<string>) => {
+  const prompt = `Write your best joke about ${ev.data}.`;
  const response = await llm.complete({ prompt });
  return new JokeEvent({ joke: response.text });
 };

-const critiqueJoke = async (_context: Context, ev: JokeEvent) => {
+const critiqueJoke = async (_: unknown, ev: JokeEvent) => {
  const prompt = `Give a thorough critique of the following joke: ${ev.data.joke}`;
  const response = await llm.complete({ prompt });
-  return new StopEvent({ result: response.text });
+  return new StopEvent(response.text);
 };

 async function validateFails() {
  try {
-    const jokeFlow = new Workflow({ verbose: true, validate: true });
-    jokeFlow.addStep(StartEvent, generateJoke, { outputs: StopEvent });
-    jokeFlow.addStep(JokeEvent, critiqueJoke, { outputs: StopEvent });
-    await jokeFlow.run("pirates");
+    const jokeFlow = new Workflow();
+    jokeFlow.addStep(
+      {
+        inputs: [StartEvent<string>],
+        outputs: [StopEvent<string>],
+      },
+      // @ts-expect-error outputs should be JokeEvent
+      generateJoke,
+    );
+    jokeFlow.addStep(
+      {
+        inputs: [JokeEvent],
+        outputs: [StopEvent],
+      },
+      critiqueJoke,
+    );
+    await jokeFlow.run("pirates").strict();
  } catch (e) {
    console.error("Validation failed:", e);
  }
 }

 async function validate() {
-  const jokeFlow = new Workflow({ verbose: true, validate: true });
-  jokeFlow.addStep(StartEvent, generateJoke, { outputs: JokeEvent });
-  jokeFlow.addStep(JokeEvent, critiqueJoke, { outputs: StopEvent });
-  const result = await jokeFlow.run("pirates");
-  console.log(result.data.result);
+  const jokeFlow = new Workflow();
+  jokeFlow.addStep(
+    {
+      inputs: [StartEvent<string>],
+      outputs: [JokeEvent],
+    },
+    generateJoke,
+  );
+  jokeFlow.addStep(
+    {
+      inputs: [JokeEvent],
+      outputs: [StopEvent<string>],
+    },
+    critiqueJoke,
+  );
+  const result = await jokeFlow.run("pirates").strict();
+  console.log(result.data);
 }

 // Usage
 async function main() {
  await validateFails();
+  console.log("---");
  await validate();
 }

@@ -1,5 +1,30 @@
 # @llamaindex/autotool

+## 5.0.13
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 5.0.12
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 5.0.11
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 5.0.10
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 5.0.9

 ### Patch Changes
@@ -1,5 +1,34 @@
 # @llamaindex/autotool-01-node-example

+## 0.0.56
+
+### Patch Changes
+
+- llamaindex@0.8.13
+- @llamaindex/autotool@5.0.13
+
+## 0.0.55
+
+### Patch Changes
+
+- llamaindex@0.8.12
+- @llamaindex/autotool@5.0.12
+
+## 0.0.54
+
+### Patch Changes
+
+- llamaindex@0.8.11
+- @llamaindex/autotool@5.0.11
+
+## 0.0.53
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+  - @llamaindex/autotool@5.0.10
+
 ## 0.0.52

 ### Patch Changes
@@ -13,5 +13,5 @@
  "scripts": {
    "start": "node --import tsx --import @llamaindex/autotool/node ./src/index.ts"
  },
-  "version": "0.0.52"
+  "version": "0.0.56"
 }
@@ -1,5 +1,34 @@
 # @llamaindex/autotool-02-next-example

+## 0.1.100
+
+### Patch Changes
+
+- llamaindex@0.8.13
+- @llamaindex/autotool@5.0.13
+
+## 0.1.99
+
+### Patch Changes
+
+- llamaindex@0.8.12
+- @llamaindex/autotool@5.0.12
+
+## 0.1.98
+
+### Patch Changes
+
+- llamaindex@0.8.11
+- @llamaindex/autotool@5.0.11
+
+## 0.1.97
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+  - @llamaindex/autotool@5.0.10
+
 ## 0.1.96

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/autotool-02-next-example",
  "private": true,
-  "version": "0.1.96",
+  "version": "0.1.100",
  "scripts": {
    "dev": "next dev",
    "build": "next build",
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/autotool",
  "type": "module",
-  "version": "5.0.9",
+  "version": "5.0.13",
  "description": "auto transpile your JS function to LLM Agent compatible",
  "files": [
    "dist",
@@ -1,5 +1,35 @@
 # @llamaindex/cloud

+## 2.0.10
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 2.0.9
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 2.0.8
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 2.0.7
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 2.0.6

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/cloud",
-  "version": "2.0.6",
+  "version": "2.0.10",
  "type": "module",
  "license": "MIT",
  "scripts": {
@@ -1,5 +1,35 @@
 # @llamaindex/community

+## 0.0.68
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.0.67
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.0.66
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.0.65
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.0.64

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/community",
  "description": "Community package for LlamaIndexTS",
-  "version": "0.0.64",
+  "version": "0.0.68",
  "type": "module",
  "types": "dist/type/index.d.ts",
  "main": "dist/cjs/index.js",
@@ -1,5 +1,31 @@
 # @llamaindex/core

+## 0.4.10
+
+### Patch Changes
+
+- a7b0ac3: fix: update tool call llm type
+- c69605f: feat: add async support to BaseChatStore and BaseChatStoreMemory
+
+## 0.4.9
+
+### Patch Changes
+
+- 7ae6eaa: feat: allow pass `additionalChatOptions` to agent
+
+## 0.4.8
+
+### Patch Changes
+
+- f865c98: feat: async get message on chat store
+
+## 0.4.7
+
+### Patch Changes
+
+- d89ebe0: feat: better support for zod schema
+- fd8c882: chore: add warning on legacy workflow API
+
 ## 0.4.6

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/core",
  "type": "module",
-  "version": "0.4.6",
+  "version": "0.4.10",
  "description": "LlamaIndex Core Module",
  "exports": {
    "./agent": {
@@ -106,11 +106,17 @@ export type AgentRunnerParams<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > = {
  llm: AI;
  chatHistory: ChatMessage<AdditionalMessageOptions>[];
  systemPrompt: MessageContent | null;
-  runner: AgentWorker<AI, Store, AdditionalMessageOptions>;
+  runner: AgentWorker<
+    AI,
+    Store,
+    AdditionalMessageOptions,
+    AdditionalChatOptions
+  >;
  tools:
    | BaseToolWithCall[]
    | ((query: MessageContent) => Promise<BaseToolWithCall[]>);
@@ -125,6 +131,7 @@ export type AgentParamsBase<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > =
  | {
      llm?: AI;
@@ -132,6 +139,7 @@ export type AgentParamsBase<
      systemPrompt?: MessageContent;
      verbose?: boolean;
      tools: BaseToolWithCall[];
+      additionalChatOptions?: AdditionalChatOptions;
    }
  | {
      llm?: AI;
@@ -139,6 +147,7 @@ export type AgentParamsBase<
      systemPrompt?: MessageContent;
      verbose?: boolean;
      toolRetriever: ObjectRetriever<BaseToolWithCall>;
+      additionalChatOptions?: AdditionalChatOptions;
    };

 /**
@@ -153,21 +162,36 @@ export abstract class AgentWorker<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > {
-  #taskSet = new Set<TaskStep<AI, Store, AdditionalMessageOptions>>();
-  abstract taskHandler: TaskHandler<AI, Store, AdditionalMessageOptions>;
+  #taskSet = new Set<
+    TaskStep<AI, Store, AdditionalMessageOptions, AdditionalChatOptions>
+  >();
+  abstract taskHandler: TaskHandler<
+    AI,
+    Store,
+    AdditionalMessageOptions,
+    AdditionalChatOptions
+  >;

  public createTask(
    query: MessageContent,
-    context: AgentTaskContext<AI, Store, AdditionalMessageOptions>,
-  ): ReadableStream<TaskStepOutput<AI, Store, AdditionalMessageOptions>> {
+    context: AgentTaskContext<
+      AI,
+      Store,
+      AdditionalMessageOptions,
+      AdditionalChatOptions
+    >,
+  ): ReadableStream<
+    TaskStepOutput<AI, Store, AdditionalMessageOptions, AdditionalChatOptions>
+  > {
    context.store.messages.push({
      role: "user",
      content: query,
    });
    const taskOutputStream = createTaskOutputStream(this.taskHandler, context);
    return new ReadableStream<
-      TaskStepOutput<AI, Store, AdditionalMessageOptions>
+      TaskStepOutput<AI, Store, AdditionalMessageOptions, AdditionalChatOptions>
    >({
      start: async (controller) => {
        for await (const stepOutput of taskOutputStream) {
@@ -176,7 +200,8 @@ export abstract class AgentWorker<
            let currentStep: TaskStep<
              AI,
              Store,
-              AdditionalMessageOptions
+              AdditionalMessageOptions,
+              AdditionalChatOptions
            > | null = stepOutput.taskStep;
            while (currentStep) {
              this.#taskSet.delete(currentStep);
@@ -227,6 +252,7 @@ export abstract class AgentRunner<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > extends BaseChatEngine {
  readonly #llm: AI;
  readonly #tools:
@@ -234,7 +260,12 @@ export abstract class AgentRunner<
    | ((query: MessageContent) => Promise<BaseToolWithCall[]>);
  readonly #systemPrompt: MessageContent | null = null;
  #chatHistory: ChatMessage<AdditionalMessageOptions>[];
-  readonly #runner: AgentWorker<AI, Store, AdditionalMessageOptions>;
+  readonly #runner: AgentWorker<
+    AI,
+    Store,
+    AdditionalMessageOptions,
+    AdditionalChatOptions
+  >;
  readonly #verbose: boolean;

  // create extra store
@@ -245,7 +276,7 @@ export abstract class AgentRunner<
  }

  static defaultTaskHandler: TaskHandler<LLM> = async (step, enqueueOutput) => {
-    const { llm, getTools, stream } = step.context;
+    const { llm, getTools, stream, additionalChatOptions } = step.context;
    const lastMessage = step.context.store.messages.at(-1)!.content;
    const tools = await getTools(lastMessage);
    if (!stream) {
@@ -253,8 +284,9 @@ export abstract class AgentRunner<
        stream,
        tools,
        messages: [...step.context.store.messages],
+        additionalChatOptions,
      });
-      await stepTools<LLM>({
+      await stepTools({
        response,
        tools,
        step,
@@ -265,6 +297,7 @@ export abstract class AgentRunner<
        stream,
        tools,
        messages: [...step.context.store.messages],
+        additionalChatOptions,
      });
      await stepToolsStreaming<LLM>({
        response,
@@ -276,7 +309,12 @@ export abstract class AgentRunner<
  };

  protected constructor(
-    params: AgentRunnerParams<AI, Store, AdditionalMessageOptions>,
+    params: AgentRunnerParams<
+      AI,
+      Store,
+      AdditionalMessageOptions,
+      AdditionalChatOptions
+    >,
  ) {
    super();
    const { llm, chatHistory, systemPrompt, runner, tools, verbose } = params;
@@ -330,6 +368,7 @@ export abstract class AgentRunner<
    stream: boolean = false,
    verbose: boolean | undefined = undefined,
    chatHistory?: ChatMessage<AdditionalMessageOptions>[],
+    additionalChatOptions?: AdditionalChatOptions,
  ) {
    const initialMessages = [...(chatHistory ?? this.#chatHistory)];
    if (this.#systemPrompt !== null) {
@@ -348,6 +387,7 @@ export abstract class AgentRunner<
      stream,
      toolCallCount: 0,
      llm: this.#llm,
+      additionalChatOptions: additionalChatOptions ?? {},
      getTools: (message) => this.getTools(message),
      store: {
        ...this.createStore(),
@@ -365,13 +405,29 @@ export abstract class AgentRunner<
    });
  }

-  async chat(params: NonStreamingChatEngineParams): Promise<EngineResponse>;
  async chat(
-    params: StreamingChatEngineParams,
+    params: NonStreamingChatEngineParams<
+      AdditionalMessageOptions,
+      AdditionalChatOptions
+    >,
+  ): Promise<EngineResponse>;
+  async chat(
+    params: StreamingChatEngineParams<
+      AdditionalMessageOptions,
+      AdditionalChatOptions
+    >,
  ): Promise<ReadableStream<EngineResponse>>;
  @wrapEventCaller
  async chat(
-    params: NonStreamingChatEngineParams | StreamingChatEngineParams,
+    params:
+      | NonStreamingChatEngineParams<
+          AdditionalMessageOptions,
+          AdditionalChatOptions
+        >
+      | StreamingChatEngineParams<
+          AdditionalMessageOptions,
+          AdditionalChatOptions
+        >,
  ): Promise<EngineResponse | ReadableStream<EngineResponse>> {
    let chatHistory: ChatMessage<AdditionalMessageOptions>[] = [];

@@ -388,6 +444,7 @@ export abstract class AgentRunner<
      !!params.stream,
      false,
      chatHistory,
+      params.chatOptions,
    );
    for await (const stepOutput of task) {
      // update chat history for each round
@@ -4,24 +4,66 @@ import { ObjectRetriever } from "../objects";
 import { AgentRunner, AgentWorker, type AgentParamsBase } from "./base.js";
 import { validateAgentParams } from "./utils.js";

-type LLMParamsBase = AgentParamsBase<LLM>;
+type LLMParamsBase<
+  AI extends LLM,
+  AdditionalMessageOptions extends object = AI extends LLM<
+    object,
+    infer AdditionalMessageOptions
+  >
+    ? AdditionalMessageOptions
+    : never,
+  AdditionalChatOptions extends object = object,
+> = AgentParamsBase<AI, AdditionalMessageOptions, AdditionalChatOptions>;

-type LLMParamsWithTools = LLMParamsBase & {
+type LLMParamsWithTools<
+  AI extends LLM,
+  AdditionalMessageOptions extends object = AI extends LLM<
+    object,
+    infer AdditionalMessageOptions
+  >
+    ? AdditionalMessageOptions
+    : never,
+  AdditionalChatOptions extends object = object,
+> = LLMParamsBase<AI, AdditionalMessageOptions, AdditionalChatOptions> & {
  tools: BaseToolWithCall[];
 };

-type LLMParamsWithToolRetriever = LLMParamsBase & {
+type LLMParamsWithToolRetriever<
+  AI extends LLM,
+  AdditionalMessageOptions extends object = AI extends LLM<
+    object,
+    infer AdditionalMessageOptions
+  >
+    ? AdditionalMessageOptions
+    : never,
+  AdditionalChatOptions extends object = object,
+> = LLMParamsBase<AI, AdditionalMessageOptions, AdditionalChatOptions> & {
  toolRetriever: ObjectRetriever<BaseToolWithCall>;
 };

-export type LLMAgentParams = LLMParamsWithTools | LLMParamsWithToolRetriever;
+export type LLMAgentParams<
+  AI extends LLM,
+  AdditionalMessageOptions extends object = AI extends LLM<
+    object,
+    infer AdditionalMessageOptions
+  >
+    ? AdditionalMessageOptions
+    : never,
+  AdditionalChatOptions extends object = object,
+> =
+  | LLMParamsWithTools<AI, AdditionalMessageOptions, AdditionalChatOptions>
+  | LLMParamsWithToolRetriever<
+      AI,
+      AdditionalMessageOptions,
+      AdditionalChatOptions
+    >;

 export class LLMAgentWorker extends AgentWorker<LLM> {
  taskHandler = AgentRunner.defaultTaskHandler;
 }

 export class LLMAgent extends AgentRunner<LLM> {
-  constructor(params: LLMAgentParams) {
+  constructor(params: LLMAgentParams<LLM>) {
    validateAgentParams(params);
    const llm = params.llm ?? (Settings.llm ? (Settings.llm as LLM) : null);
    if (!llm)
@@ -19,6 +19,7 @@ export type AgentTaskContext<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > = {
  readonly stream: boolean;
  readonly toolCallCount: number;
@@ -26,6 +27,7 @@ export type AgentTaskContext<
  readonly getTools: (
    input: MessageContent,
  ) => BaseToolWithCall[] | Promise<BaseToolWithCall[]>;
+  readonly additionalChatOptions: Partial<AdditionalChatOptions>;
  shouldContinue: (
    taskStep: Readonly<TaskStep<Model, Store, AdditionalMessageOptions>>,
  ) => boolean;
@@ -45,13 +47,26 @@ export type TaskStep<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > = {
  id: UUID;
-  context: AgentTaskContext<Model, Store, AdditionalMessageOptions>;
+  context: AgentTaskContext<
+    Model,
+    Store,
+    AdditionalMessageOptions,
+    AdditionalChatOptions
+  >;

  // linked list
-  prevStep: TaskStep<Model, Store, AdditionalMessageOptions> | null;
-  nextSteps: Set<TaskStep<Model, Store, AdditionalMessageOptions>>;
+  prevStep: TaskStep<
+    Model,
+    Store,
+    AdditionalMessageOptions,
+    AdditionalChatOptions
+  > | null;
+  nextSteps: Set<
+    TaskStep<Model, Store, AdditionalMessageOptions, AdditionalChatOptions>
+  >;
 };

 export type TaskStepOutput<
@@ -63,8 +78,14 @@ export type TaskStepOutput<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > = {
-  taskStep: TaskStep<Model, Store, AdditionalMessageOptions>;
+  taskStep: TaskStep<
+    Model,
+    Store,
+    AdditionalMessageOptions,
+    AdditionalChatOptions
+  >;
  // output shows the response to the user
  output:
    | ChatResponse<AdditionalMessageOptions>
@@ -81,10 +102,16 @@ export type TaskHandler<
  >
    ? AdditionalMessageOptions
    : never,
+  AdditionalChatOptions extends object = object,
 > = (
-  step: TaskStep<Model, Store, AdditionalMessageOptions>,
+  step: TaskStep<Model, Store, AdditionalMessageOptions, AdditionalChatOptions>,
  enqueueOutput: (
-    taskOutput: TaskStepOutput<Model, Store, AdditionalMessageOptions>,
+    taskOutput: TaskStepOutput<
+      Model,
+      Store,
+      AdditionalMessageOptions,
+      AdditionalChatOptions
+    >,
  ) => void,
 ) => Promise<void>;

@@ -16,14 +16,18 @@ export interface BaseChatEngineParams<

 export interface StreamingChatEngineParams<
  AdditionalMessageOptions extends object = object,
+  AdditionalChatOptions extends object = object,
 > extends BaseChatEngineParams<AdditionalMessageOptions> {
  stream: true;
+  chatOptions?: AdditionalChatOptions;
 }

 export interface NonStreamingChatEngineParams<
  AdditionalMessageOptions extends object = object,
+  AdditionalChatOptions extends object = object,
 > extends BaseChatEngineParams<AdditionalMessageOptions> {
  stream?: false;
+  chatOptions?: AdditionalChatOptions;
 }

 export abstract class BaseChatEngine {
@@ -66,6 +66,8 @@ export abstract class BaseLLM<

 export abstract class ToolCallLLM<
  AdditionalChatOptions extends object = object,
-> extends BaseLLM<AdditionalChatOptions, ToolCallLLMMessageOptions> {
+  AdditionalMessageOptions extends
+    ToolCallLLMMessageOptions = ToolCallLLMMessageOptions,
+> extends BaseLLM<AdditionalChatOptions, AdditionalMessageOptions> {
  abstract supportToolCall: boolean;
 }
@@ -65,19 +65,21 @@ export abstract class BaseChatStoreMemory<
    super();
  }

-  getAllMessages(): ChatMessage<AdditionalMessageOptions>[] {
+  getAllMessages():
+    | ChatMessage<AdditionalMessageOptions>[]
+    | Promise<ChatMessage<AdditionalMessageOptions>[]> {
    return this.chatStore.getMessages(this.chatStoreKey);
  }

-  put(messages: ChatMessage<AdditionalMessageOptions>) {
+  put(messages: ChatMessage<AdditionalMessageOptions>): void | Promise<void> {
    this.chatStore.addMessage(this.chatStoreKey, messages);
  }

-  set(messages: ChatMessage<AdditionalMessageOptions>[]) {
+  set(messages: ChatMessage<AdditionalMessageOptions>[]): void | Promise<void> {
    this.chatStore.setMessages(this.chatStoreKey, messages);
  }

-  reset() {
+  reset(): void | Promise<void> {
    this.chatStore.deleteMessages(this.chatStoreKey);
  }
 }
@@ -33,11 +33,11 @@ export class ChatMemoryBuffer<
    }
  }

-  getMessages(
+  async getMessages(
    transientMessages?: ChatMessage<AdditionalMessageOptions>[] | undefined,
    initialTokenCount: number = 0,
  ) {
-    const messages = this.getAllMessages();
+    const messages = await this.getAllMessages();

    if (initialTokenCount > this.tokenLimit) {
      throw new Error("Initial token count exceeds token limit");
@@ -7,7 +7,11 @@ export abstract class BaseChatStore<
    key: string,
    messages: ChatMessage<AdditionalMessageOptions>[],
  ): void;
-  abstract getMessages(key: string): ChatMessage<AdditionalMessageOptions>[];
+  abstract getMessages(
+    key: string,
+  ):
+    | ChatMessage<AdditionalMessageOptions>[]
+    | Promise<ChatMessage<AdditionalMessageOptions>[]>;
  abstract addMessage(
    key: string,
    message: ChatMessage<AdditionalMessageOptions>,
@@ -15,5 +19,7 @@ export abstract class BaseChatStore<
  ): void;
  abstract deleteMessages(key: string): void;
  abstract deleteMessage(key: string, idx: number): void;
-  abstract getKeys(): IterableIterator<string>;
+  abstract getKeys():
+    | IterableIterator<string>
+    | Promise<IterableIterator<string>>;
 }
@@ -4,18 +4,12 @@ import { zodToJsonSchema } from "zod-to-json-schema";
 import type { JSONValue } from "../global";
 import type { BaseTool, ToolMetadata } from "../llms";

-const kOriginalFn = Symbol("originalFn");
-
 export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
  implements BaseTool<T>
 {
-  [kOriginalFn]?: (input: T) => R;
-
  #fn: (input: T) => R;
-  #metadata: ToolMetadata<JSONSchemaType<T>>;
-  // todo: for the future, we can use zod to validate the input parameters
-  // eslint-disable-next-line no-unused-private-class-members
-  #zodType: z.ZodType<T> | null = null;
+  readonly #metadata: ToolMetadata<JSONSchemaType<T>>;
+  readonly #zodType: z.ZodType<T> | null = null;
  constructor(
    fn: (input: T) => R,
    metadata: ToolMetadata<JSONSchemaType<T>>,
@@ -32,6 +26,12 @@ export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
    fn: (input: T) => JSONValue | Promise<JSONValue>,
    schema: ToolMetadata<JSONSchemaType<T>>,
  ): FunctionTool<T, JSONValue | Promise<JSONValue>>;
+  static from<R extends z.ZodType>(
+    fn: (input: z.infer<R>) => JSONValue | Promise<JSONValue>,
+    schema: Omit<ToolMetadata, "parameters"> & {
+      parameters: R;
+    },
+  ): FunctionTool<z.infer<R>, JSONValue | Promise<JSONValue>>;
  static from<T, R extends z.ZodType<T>>(
    fn: (input: T) => JSONValue | Promise<JSONValue>,
    schema: Omit<ToolMetadata, "parameters"> & {
@@ -40,15 +40,15 @@ export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
  ): FunctionTool<T, JSONValue>;
  // eslint-disable-next-line @typescript-eslint/no-explicit-any
  static from(fn: any, schema: any): any {
-    if (schema.parameter instanceof z.ZodSchema) {
-      const jsonSchema = zodToJsonSchema(schema.parameter);
+    if (schema.parameters instanceof z.ZodSchema) {
+      const jsonSchema = zodToJsonSchema(schema.parameters);
      return new FunctionTool(
        fn,
        {
          ...schema,
          parameters: jsonSchema,
        },
-        schema.parameter,
+        schema.parameters,
      );
    }
    return new FunctionTool(fn, schema);
@@ -58,7 +58,15 @@ export class FunctionTool<T, R extends JSONValue | Promise<JSONValue>>
    return this.#metadata as BaseTool<T>["metadata"];
  }

-  call(input: T) {
+  call = (input: T) => {
+    if (this.#zodType) {
+      const result = this.#zodType.safeParse(input);
+      if (result.success) {
+        return this.#fn.call(null, result.data);
+      } else {
+        console.warn(result.error.errors);
+      }
+    }
    return this.#fn.call(null, input);
-  }
+  };
 }
@@ -13,6 +13,8 @@ export type StepFunction<T extends WorkflowEvent = WorkflowEvent> = (

 type EventTypeParam = EventTypes | EventTypes[];

+let once = false;
+
 export class Workflow {
  #steps: Map<
    // eslint-disable-next-line @typescript-eslint/no-explicit-any
@@ -29,8 +31,20 @@ export class Workflow {
      verbose?: boolean;
      timeout?: number;
      validate?: boolean;
+      ignoreDeprecatedWarning?: boolean;
    } = {},
  ) {
+    if (!once && !params.ignoreDeprecatedWarning) {
+      console.warn(
+        "@llamaindex/core/workflow is going to use the new workflow API in the next major version.",
+        "Please update your imports to @llamaindex/workflow",
+      );
+      console.warn(
+        "See https://ts.llamaindex.ai/docs/llamaindex/guide/workflow for more information",
+      );
+      once = true;
+    }
+
    this.#verbose = params.verbose ?? false;
    this.#timeout = params.timeout ?? null;
    this.#validate = params.validate ?? false;
@@ -19,7 +19,7 @@ describe("ChatMemoryBuffer", () => {
    expect(buffer.tokenLimit).toBe(500);
  });

-  test("getMessages returns all messages when under token limit", () => {
+  test("getMessages returns all messages when under token limit", async () => {
    const messages: ChatMessage[] = [
      { role: "user", content: "Hello" },
      { role: "assistant", content: "Hi there!" },
@@ -30,11 +30,11 @@ describe("ChatMemoryBuffer", () => {
      chatHistory: messages,
    });

-    const result = buffer.getMessages();
+    const result = await buffer.getMessages();
    expect(result).toEqual(messages);
  });

-  test("getMessages truncates messages when over token limit", () => {
+  test("getMessages truncates messages when over token limit", async () => {
    const messages: ChatMessage[] = [
      { role: "user", content: "This is a long message" },
      { role: "assistant", content: "This is also a long reply" },
@@ -45,11 +45,11 @@ describe("ChatMemoryBuffer", () => {
      chatHistory: messages,
    });

-    const result = buffer.getMessages();
+    const result = await buffer.getMessages();
    expect(result).toEqual([{ role: "user", content: "Short" }]);
  });

-  test("getMessages handles input messages", () => {
+  test("getMessages handles input messages", async () => {
    const storedMessages: ChatMessage[] = [
      { role: "user", content: "Hello" },
      { role: "assistant", content: "Hi there!" },
@@ -62,13 +62,13 @@ describe("ChatMemoryBuffer", () => {
    const inputMessages: ChatMessage[] = [
      { role: "user", content: "New message" },
    ];
-    const result = buffer.getMessages(inputMessages);
+    const result = await buffer.getMessages(inputMessages);
    expect(result).toEqual([...inputMessages, ...storedMessages]);
  });

  test("getMessages throws error when initial token count exceeds limit", () => {
    const buffer = new ChatMemoryBuffer({ tokenLimit: 10 });
-    expect(() => buffer.getMessages(undefined, 20)).toThrow(
+    expect(async () => buffer.getMessages(undefined, 20)).rejects.toThrow(
      "Initial token count exceeds token limit",
    );
  });
@@ -1,5 +1,30 @@
 # @llamaindex/experimental

+## 0.0.125
+
+### Patch Changes
+
+- llamaindex@0.8.13
+
+## 0.0.124
+
+### Patch Changes
+
+- llamaindex@0.8.12
+
+## 0.0.123
+
+### Patch Changes
+
+- llamaindex@0.8.11
+
+## 0.0.122
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+  - llamaindex@0.8.10
+
 ## 0.0.121

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/experimental",
  "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.121",
+  "version": "0.0.125",
  "type": "module",
  "types": "dist/type/index.d.ts",
  "main": "dist/cjs/index.js",
@@ -1,5 +1,90 @@
 # llamaindex

+## 0.8.13
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/anthropic@0.0.19
+  - @llamaindex/core@0.4.10
+  - @llamaindex/cloud@2.0.10
+  - @llamaindex/node-parser@0.0.11
+  - @llamaindex/clip@0.0.19
+  - @llamaindex/deepinfra@0.0.19
+  - @llamaindex/huggingface@0.0.19
+  - @llamaindex/ollama@0.0.26
+  - @llamaindex/openai@0.1.35
+  - @llamaindex/portkey-ai@0.0.19
+  - @llamaindex/replicate@0.0.19
+  - @llamaindex/readers@1.0.11
+  - @llamaindex/groq@0.0.34
+  - @llamaindex/vllm@0.0.5
+
+## 0.8.12
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+  - @llamaindex/openai@0.1.34
+  - @llamaindex/cloud@2.0.9
+  - @llamaindex/node-parser@0.0.10
+  - @llamaindex/anthropic@0.0.18
+  - @llamaindex/clip@0.0.18
+  - @llamaindex/deepinfra@0.0.18
+  - @llamaindex/huggingface@0.0.18
+  - @llamaindex/ollama@0.0.25
+  - @llamaindex/portkey-ai@0.0.18
+  - @llamaindex/replicate@0.0.18
+  - @llamaindex/readers@1.0.10
+  - @llamaindex/groq@0.0.33
+  - @llamaindex/vllm@0.0.4
+
+## 0.8.11
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+  - @llamaindex/cloud@2.0.8
+  - @llamaindex/node-parser@0.0.9
+  - @llamaindex/anthropic@0.0.17
+  - @llamaindex/clip@0.0.17
+  - @llamaindex/deepinfra@0.0.17
+  - @llamaindex/huggingface@0.0.17
+  - @llamaindex/ollama@0.0.24
+  - @llamaindex/openai@0.1.33
+  - @llamaindex/portkey-ai@0.0.17
+  - @llamaindex/replicate@0.0.17
+  - @llamaindex/readers@1.0.9
+  - @llamaindex/groq@0.0.32
+  - @llamaindex/vllm@0.0.3
+
+## 0.8.10
+
+### Patch Changes
+
+- f066e50: feat: vllm support
+- Updated dependencies [f066e50]
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/vllm@0.0.2
+  - @llamaindex/core@0.4.7
+  - @llamaindex/cloud@2.0.7
+  - @llamaindex/node-parser@0.0.8
+  - @llamaindex/anthropic@0.0.16
+  - @llamaindex/clip@0.0.16
+  - @llamaindex/deepinfra@0.0.16
+  - @llamaindex/huggingface@0.0.16
+  - @llamaindex/ollama@0.0.23
+  - @llamaindex/openai@0.1.32
+  - @llamaindex/portkey-ai@0.0.16
+  - @llamaindex/replicate@0.0.16
+  - @llamaindex/readers@1.0.8
+  - @llamaindex/groq@0.0.31
+
 ## 0.8.9

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "llamaindex",
-  "version": "0.8.9",
+  "version": "0.8.13",
  "license": "MIT",
  "type": "module",
  "keywords": [
@@ -44,6 +44,7 @@
    "@llamaindex/portkey-ai": "workspace:*",
    "@llamaindex/readers": "workspace:*",
    "@llamaindex/replicate": "workspace:*",
+    "@llamaindex/vllm": "workspace:*",
    "@mistralai/mistralai": "^1.0.4",
    "@mixedbread-ai/sdk": "^2.2.11",
    "@pinecone-database/pinecone": "^3.0.2",
@@ -85,10 +86,10 @@
    }
  },
  "devDependencies": {
+    "@huggingface/transformers": "^3.0.2",
    "@swc/cli": "^0.5.0",
    "@swc/core": "^1.7.22",
    "@vercel/postgres": "^0.10.0",
-    "@huggingface/transformers": "^3.0.2",
    "concurrently": "^9.1.0",
    "glob": "^11.0.0",
    "pg": "^8.12.0",
@@ -1,3 +1,4 @@
+export { VLLM, type VLLMParams } from "@llamaindex/vllm";
 export {
  ALL_AVAILABLE_ANTHROPIC_LEGACY_MODELS,
  ALL_AVAILABLE_ANTHROPIC_MODELS,
@@ -6,7 +7,6 @@ export {
 } from "./anthropic.js";
 export { FireworksLLM } from "./fireworks.js";
 export { Gemini, GeminiSession } from "./gemini/base.js";
-
 export {
  GEMINI_MODEL,
  type GoogleGeminiSessionOptions,
@@ -1,5 +1,35 @@
 # @llamaindex/node-parser

+## 0.0.11
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.0.10
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.0.9
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.0.8
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.0.7

 ### Patch Changes
@@ -1,6 +1,6 @@
 {
  "name": "@llamaindex/node-parser",
-  "version": "0.0.7",
+  "version": "0.0.11",
  "description": "Node parser for LlamaIndex",
  "type": "module",
  "exports": {
@@ -1,5 +1,36 @@
 # @llamaindex/anthropic

+## 0.0.19
+
+### Patch Changes
+
+- a7b0ac3: feat(anthropic): support prompt caching
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.0.15

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/anthropic",
  "description": "Anthropic Adapter for LlamaIndex",
-  "version": "0.0.15",
+  "version": "0.0.19",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -33,7 +33,7 @@
    "bunchee": "5.6.1"
  },
  "dependencies": {
-    "@anthropic-ai/sdk": "0.27.1",
+    "@anthropic-ai/sdk": "0.32.1",
    "@llamaindex/core": "workspace:*",
    "@llamaindex/env": "workspace:*",
    "remeda": "^2.12.0"
@@ -11,7 +11,7 @@ import { Settings } from "@llamaindex/core/global";
 import type { EngineResponse } from "@llamaindex/core/schema";
 import { Anthropic } from "./llm.js";

-export type AnthropicAgentParams = LLMAgentParams;
+export type AnthropicAgentParams = LLMAgentParams<Anthropic>;

 export class AnthropicAgentWorker extends LLMAgentWorker {}

@@ -1,5 +1,9 @@
 import type { ClientOptions } from "@anthropic-ai/sdk";
 import { Anthropic as SDKAnthropic } from "@anthropic-ai/sdk";
+import type {
+  BetaCacheControlEphemeral,
+  BetaTextBlockParam,
+} from "@anthropic-ai/sdk/resources/beta/index";
 import type {
  TextBlock,
  TextBlockParam,
@@ -8,6 +12,7 @@ import type {
  ImageBlockParam,
  MessageCreateParamsNonStreaming,
  MessageParam,
+  Model,
  Tool,
  ToolResultBlockParam,
  ToolUseBlock,
@@ -75,6 +80,9 @@ export const ALL_AVAILABLE_ANTHROPIC_LEGACY_MODELS = {
  "claude-2.1": {
    contextWindow: 200000,
  },
+  "claude-2.0": {
+    contextWindow: 100000,
+  },
  "claude-instant-1.2": {
    contextWindow: 100000,
  },
@@ -82,18 +90,30 @@ export const ALL_AVAILABLE_ANTHROPIC_LEGACY_MODELS = {

 export const ALL_AVAILABLE_V3_MODELS = {
  "claude-3-opus": { contextWindow: 200000 },
+  "claude-3-opus-latest": { contextWindow: 200000 },
+  "claude-3-opus-20240229": { contextWindow: 200000 },
  "claude-3-sonnet": { contextWindow: 200000 },
+  "claude-3-sonnet-20240229": { contextWindow: 200000 },
  "claude-3-haiku": { contextWindow: 200000 },
+  "claude-3-haiku-20240307": { contextWindow: 200000 },
 };

 export const ALL_AVAILABLE_V3_5_MODELS = {
  "claude-3-5-sonnet": { contextWindow: 200000 },
+  "claude-3-5-sonnet-20241022": { contextWindow: 200000 },
+  "claude-3-5-sonnet-20240620": { contextWindow: 200000 },
+  "claude-3-5-sonnet-latest": { contextWindow: 200000 },
+  "claude-3-5-haiku": { contextWindow: 200000 },
+  "claude-3-5-haiku-latest": { contextWindow: 200000 },
+  "claude-3-5-haiku-20241022": { contextWindow: 200000 },
 };

 export const ALL_AVAILABLE_ANTHROPIC_MODELS = {
  ...ALL_AVAILABLE_ANTHROPIC_LEGACY_MODELS,
  ...ALL_AVAILABLE_V3_MODELS,
  ...ALL_AVAILABLE_V3_5_MODELS,
+} satisfies {
+  [key in Model]: { contextWindow: number };
 };

 const AVAILABLE_ANTHROPIC_MODELS_WITHOUT_DATE: { [key: string]: string } = {
@@ -104,10 +124,16 @@ const AVAILABLE_ANTHROPIC_MODELS_WITHOUT_DATE: { [key: string]: string } = {
 } as { [key in keyof typeof ALL_AVAILABLE_ANTHROPIC_MODELS]: string };

 export type AnthropicAdditionalChatOptions = object;
+export type AnthropicToolCallLLMMessageOptions = ToolCallLLMMessageOptions & {
+  cache_control?: BetaCacheControlEphemeral | null;
+};

-export class Anthropic extends ToolCallLLM<AnthropicAdditionalChatOptions> {
+export class Anthropic extends ToolCallLLM<
+  AnthropicAdditionalChatOptions,
+  AnthropicToolCallLLMMessageOptions
+> {
  // Per completion Anthropic params
-  model: keyof typeof ALL_AVAILABLE_ANTHROPIC_MODELS;
+  model: keyof typeof ALL_AVAILABLE_ANTHROPIC_MODELS | ({} & string);
  temperature: number;
  topP: number;
  maxTokens?: number | undefined;
@@ -147,7 +173,12 @@ export class Anthropic extends ToolCallLLM<AnthropicAdditionalChatOptions> {
      temperature: this.temperature,
      topP: this.topP,
      maxTokens: this.maxTokens,
-      contextWindow: ALL_AVAILABLE_ANTHROPIC_MODELS[this.model].contextWindow,
+      contextWindow:
+        this.model in ALL_AVAILABLE_ANTHROPIC_MODELS
+          ? ALL_AVAILABLE_ANTHROPIC_MODELS[
+              this.model as keyof typeof ALL_AVAILABLE_ANTHROPIC_MODELS
+            ].contextWindow
+          : 200000,
      tokenizer: undefined,
    };
  }
@@ -291,56 +322,74 @@ export class Anthropic extends ToolCallLLM<AnthropicAdditionalChatOptions> {
  chat(
    params: LLMChatParamsStreaming<
      AnthropicAdditionalChatOptions,
-      ToolCallLLMMessageOptions
+      AnthropicToolCallLLMMessageOptions
    >,
-  ): Promise<AsyncIterable<ChatResponseChunk<ToolCallLLMMessageOptions>>>;
+  ): Promise<
+    AsyncIterable<ChatResponseChunk<AnthropicToolCallLLMMessageOptions>>
+  >;
  chat(
    params: LLMChatParamsNonStreaming<
      AnthropicAdditionalChatOptions,
-      ToolCallLLMMessageOptions
+      AnthropicToolCallLLMMessageOptions
    >,
-  ): Promise<ChatResponse<ToolCallLLMMessageOptions>>;
+  ): Promise<ChatResponse<AnthropicToolCallLLMMessageOptions>>;
  @wrapLLMEvent
  async chat(
    params:
      | LLMChatParamsNonStreaming<
          AnthropicAdditionalChatOptions,
-          ToolCallLLMMessageOptions
+          AnthropicToolCallLLMMessageOptions
        >
      | LLMChatParamsStreaming<
          AnthropicAdditionalChatOptions,
-          ToolCallLLMMessageOptions
+          AnthropicToolCallLLMMessageOptions
        >,
  ): Promise<
-    | ChatResponse<ToolCallLLMMessageOptions>
-    | AsyncIterable<ChatResponseChunk<ToolCallLLMMessageOptions>>
+    | ChatResponse<AnthropicToolCallLLMMessageOptions>
+    | AsyncIterable<ChatResponseChunk<AnthropicToolCallLLMMessageOptions>>
  > {
    let { messages } = params;

    const { stream, tools } = params;

-    let systemPrompt: string | null = null;
+    let systemPrompt: string | Array<BetaTextBlockParam> | null = null;

    const systemMessages = messages.filter(
      (message) => message.role === "system",
    );

    if (systemMessages.length > 0) {
-      systemPrompt = systemMessages
-        .map((message) => message.content)
-        .join("\n");
+      systemPrompt = systemMessages.map((message) =>
+        message.options && "cache_control" in message.options
+          ? {
+              type: "text",
+              text: extractText(message.content),
+              cache_control: message.options.cache_control,
+            }
+          : {
+              type: "text",
+              text: extractText(message.content),
+            },
+      );
      messages = messages.filter((message) => message.role !== "system");
    }
+    const beta =
+      systemPrompt?.find((message) => "cache_control" in message) !== undefined;
+
+    // case: Non-streaming
+    let anthropic = this.session.anthropic;
+    if (beta) {
+      // @ts-expect-error type casting
+      anthropic = anthropic.beta.promptCaching;
+    }

    // case: Streaming
    if (stream) {
      if (tools) {
        console.error("Tools are not supported in streaming mode");
      }
-      return this.streamChat(messages, systemPrompt);
+      return this.streamChat(messages, systemPrompt, anthropic);
    }
-    // case: Non-streaming
-    const anthropic = this.session.anthropic;

    if (tools) {
      const params: MessageCreateParamsNonStreaming = {
@@ -378,7 +427,10 @@ export class Anthropic extends ToolCallLLM<AnthropicAdditionalChatOptions> {
                  toolCall: toolUseBlock.map((block) => ({
                    id: block.id,
                    name: block.name,
-                    input: block.input,
+                    input:
+                      typeof block.input === "object"
+                        ? JSON.stringify(block.input)
+                        : `${block.input}`,
                  })),
                }
              : {},
@@ -411,10 +463,11 @@ export class Anthropic extends ToolCallLLM<AnthropicAdditionalChatOptions> {
  }

  protected async *streamChat(
-    messages: ChatMessage<ToolCallLLMMessageOptions>[],
-    systemPrompt?: string | null,
-  ): AsyncIterable<ChatResponseChunk<ToolCallLLMMessageOptions>> {
-    const stream = await this.session.anthropic.messages.create({
+    messages: ChatMessage<AnthropicToolCallLLMMessageOptions>[],
+    systemPrompt: string | Array<BetaTextBlockParam> | null,
+    anthropic: SDKAnthropic,
+  ): AsyncIterable<ChatResponseChunk<AnthropicToolCallLLMMessageOptions>> {
+    const stream = await anthropic.messages.create({
      model: this.getModelName(this.model),
      messages: this.formatMessages(messages),
      max_tokens: this.maxTokens ?? 4096,
@@ -1,5 +1,39 @@
 # @llamaindex/clip

+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+  - @llamaindex/openai@0.1.35
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+  - @llamaindex/openai@0.1.34
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+  - @llamaindex/openai@0.1.33
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+  - @llamaindex/openai@0.1.32
+
 ## 0.0.15

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/clip",
  "description": "Clip Embedding Adapter for LlamaIndex",
-  "version": "0.0.15",
+  "version": "0.0.19",
  "type": "module",
  "types": "dist/index.d.ts",
  "main": "dist/index.cjs",
@@ -1,5 +1,39 @@
 # @llamaindex/deepinfra

+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+  - @llamaindex/openai@0.1.35
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+  - @llamaindex/openai@0.1.34
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+  - @llamaindex/openai@0.1.33
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+  - @llamaindex/openai@0.1.32
+
 ## 0.0.15

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/deepinfra",
  "description": "Deepinfra Adapter for LlamaIndex",
-  "version": "0.0.15",
+  "version": "0.0.19",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -1,5 +1,30 @@
 # @llamaindex/groq

+## 0.0.34
+
+### Patch Changes
+
+- @llamaindex/openai@0.1.35
+
+## 0.0.33
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/openai@0.1.34
+
+## 0.0.32
+
+### Patch Changes
+
+- @llamaindex/openai@0.1.33
+
+## 0.0.31
+
+### Patch Changes
+
+- @llamaindex/openai@0.1.32
+
 ## 0.0.30

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/groq",
  "description": "Groq Adapter for LlamaIndex",
-  "version": "0.0.30",
+  "version": "0.0.34",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -1,5 +1,39 @@
 # @llamaindex/huggingface

+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+  - @llamaindex/openai@0.1.35
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+  - @llamaindex/openai@0.1.34
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+  - @llamaindex/openai@0.1.33
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+  - @llamaindex/openai@0.1.32
+
 ## 0.0.15

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/huggingface",
  "description": "Huggingface Adapter for LlamaIndex",
-  "version": "0.0.15",
+  "version": "0.0.19",
  "type": "module",
  "types": "dist/index.d.ts",
  "main": "dist/index.cjs",
@@ -1,5 +1,35 @@
 # @llamaindex/ollama

+## 0.0.26
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.0.25
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.0.24
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.0.23
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.0.22

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/ollama",
  "description": "Ollama Adapter for LlamaIndex",
-  "version": "0.0.22",
+  "version": "0.0.26",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -8,7 +8,7 @@ import { Ollama } from "./llm";

 // This is likely not necessary anymore but leaving it here just incase it's in use elsewhere

-export type OllamaAgentParams = LLMAgentParams & {
+export type OllamaAgentParams = LLMAgentParams<Ollama> & {
  model?: string;
 };

@@ -1,5 +1,36 @@
 # @llamaindex/openai

+## 0.1.35
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.1.34
+
+### Patch Changes
+
+- 7ae6eaa: feat: allow pass `additionalChatOptions` to agent
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.1.33
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.1.32
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.1.31

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/openai",
  "description": "OpenAI Adapter for LlamaIndex",
-  "version": "0.1.31",
+  "version": "0.1.35",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -35,7 +35,6 @@
  "dependencies": {
    "@llamaindex/core": "workspace:*",
    "@llamaindex/env": "workspace:*",
-    "openai": "^4.68.1",
-    "remeda": "^2.12.0"
+    "openai": "^4.68.1"
  }
 }
@@ -4,11 +4,16 @@ import {
  type LLMAgentParams,
 } from "@llamaindex/core/agent";
 import { Settings } from "@llamaindex/core/global";
-import { OpenAI } from "./llm";
+import type { ToolCallLLMMessageOptions } from "@llamaindex/core/llms";
+import { OpenAI, type OpenAIAdditionalChatOptions } from "./llm";

-// This is likely not necessary anymore but leaving it here just incase it's in use elsewhere
+// This is likely not necessary anymore but leaving it here just in case it's in use elsewhere

-export type OpenAIAgentParams = LLMAgentParams;
+export type OpenAIAgentParams = LLMAgentParams<
+  OpenAI,
+  ToolCallLLMMessageOptions,
+  OpenAIAdditionalChatOptions
+>;

 export class OpenAIAgentWorker extends LLMAgentWorker {}

@@ -1,5 +1,35 @@
 # @llamaindex/portkey-ai

+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.0.15

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/portkey-ai",
  "description": "Portkey Adapter for LlamaIndex",
-  "version": "0.0.15",
+  "version": "0.0.19",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -1,5 +1,35 @@
 # @llamaindex/replicate

+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 0.0.15

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/replicate",
  "description": "Replicate Adapter for LlamaIndex",
-  "version": "0.0.15",
+  "version": "0.0.19",
  "type": "module",
  "main": "./dist/index.cjs",
  "module": "./dist/index.js",
@@ -0,0 +1,27 @@
+# @llamaindex/vllm
+
+## 0.0.5
+
+### Patch Changes
+
+- @llamaindex/openai@0.1.35
+
+## 0.0.4
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/openai@0.1.34
+
+## 0.0.3
+
+### Patch Changes
+
+- @llamaindex/openai@0.1.33
+
+## 0.0.2
+
+### Patch Changes
+
+- f066e50: feat: vllm support
+  - @llamaindex/openai@0.1.32
@@ -0,0 +1,38 @@
+{
+  "name": "@llamaindex/vllm",
+  "description": "vLLM Adapter for LlamaIndex",
+  "version": "0.0.5",
+  "type": "module",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
+  "exports": {
+    ".": {
+      "require": {
+        "types": "./dist/index.d.cts",
+        "default": "./dist/index.cjs"
+      },
+      "import": {
+        "types": "./dist/index.d.ts",
+        "default": "./dist/index.js"
+      }
+    }
+  },
+  "files": [
+    "dist"
+  ],
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "directory": "packages/providers/vllm"
+  },
+  "scripts": {
+    "build": "bunchee",
+    "dev": "bunchee --watch"
+  },
+  "devDependencies": {
+    "bunchee": "5.6.1"
+  },
+  "dependencies": {
+    "@llamaindex/openai": "workspace:*"
+  }
+}
@@ -0,0 +1 @@
+export { VLLM, type VLLMParams } from "./llm";
@@ -0,0 +1,25 @@
+/**
+ * vLLM
+ *
+ * https://docs.vllm.ai/en/latest/index.html
+ *
+ * @module
+ */
+import { OpenAI } from "@llamaindex/openai";
+
+export type VLLMParams = {
+  model: string;
+  baseURL?: string;
+};
+
+export class VLLM extends OpenAI {
+  constructor(params: VLLMParams) {
+    super({
+      additionalSessionOptions: {
+        baseURL: "http://localhost:8000/v1",
+      },
+      model: params.model,
+      apiKey: "token-abc123",
+    });
+  }
+}
@@ -0,0 +1,16 @@
+{
+  "extends": "../../../tsconfig.json",
+  "compilerOptions": {
+    "target": "ESNext",
+    "module": "ESNext",
+    "moduleResolution": "bundler",
+    "outDir": "./lib",
+    "tsBuildInfoFile": "./lib/.tsbuildinfo"
+  },
+  "include": ["./src", "package.json"],
+  "references": [
+    {
+      "path": "../openai/tsconfig.json"
+    }
+  ]
+}
@@ -1,5 +1,35 @@
 # @llamaindex/readers

+## 1.0.11
+
+### Patch Changes
+
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+
+## 1.0.10
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+
+## 1.0.9
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+
+## 1.0.8
+
+### Patch Changes
+
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+  - @llamaindex/core@0.4.7
+
 ## 1.0.7

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/readers",
  "description": "LlamaIndex Readers",
-  "version": "1.0.7",
+  "version": "1.0.11",
  "type": "module",
  "exports": {
    "./node/hook": "./node/dist/hook.js",
@@ -1,5 +1,24 @@
 # @llamaindex/workflow

+## 0.0.6
+
+### Patch Changes
+
+- ee20c44: feat(workflow): allow send event with no output
+
+## 0.0.5
+
+### Patch Changes
+
+- ea92b69: fix: output event check
+- fadc8b8: feat: recoverable context with error handling
+
+## 0.0.4
+
+### Patch Changes
+
+- fd8c882: fix: infinite loop issue caused by multiple inputs in edge cases
+
 ## 0.0.3

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/workflow",
  "description": "Workflow API",
-  "version": "0.0.3",
+  "version": "0.0.6",
  "type": "module",
  "types": "dist/index.d.ts",
  "module": "dist/index.js",
@@ -13,19 +13,18 @@ export type StepHandler<
    AnyWorkflowEventConstructor | StartEventConstructor,
    ...(AnyWorkflowEventConstructor | StopEventConstructor)[],
  ] = [AnyWorkflowEventConstructor | StartEventConstructor],
-  Out extends [
-    AnyWorkflowEventConstructor | StartEventConstructor,
-    ...(AnyWorkflowEventConstructor | StopEventConstructor)[],
-  ] = [AnyWorkflowEventConstructor | StopEventConstructor],
+  Out extends (AnyWorkflowEventConstructor | StopEventConstructor)[] = [],
 > = (
  context: HandlerContext<Data>,
  ...events: {
    [K in keyof Inputs]: InstanceType<Inputs[K]>;
  }
 ) => Promise<
-  {
-    [K in keyof Out]: InstanceType<Out[K]>;
-  }[number]
+  Out extends []
+    ? void
+    : {
+        [K in keyof Out]: InstanceType<Out[K]>;
+      }[number]
 >;

 export type ReadonlyStepMap<Data> = ReadonlyMap<
@@ -275,7 +274,7 @@ export class WorkflowContext<Start = string, Stop = string, Data = unknown>
   */
  #createStreamEvents(): AsyncIterableIterator<WorkflowEvent<unknown>> {
    const isPendingEvents = new WeakSet<WorkflowEvent<unknown>>();
-    const pendingTasks = new Set<Promise<WorkflowEvent<unknown>>>();
+    const pendingTasks = new Set<Promise<WorkflowEvent<unknown> | void>>();
    const enqueuedEvents = new Set<WorkflowEvent<unknown>>();
    const stream = new ReadableStream<WorkflowEvent<unknown>>({
      start: async (controller) => {
@@ -289,6 +288,15 @@ export class WorkflowContext<Start = string, Stop = string, Data = unknown>
                  (event) => event instanceof requestEvent,
                );
                if (acceptableInput) {
+                  // remove the event from the queue, in case of infinite loop
+                  const protocolIdx = this.#queue.findIndex(
+                    (protocol) =>
+                      protocol.type === "event" &&
+                      protocol.event === acceptableInput,
+                  );
+                  if (protocolIdx !== -1) {
+                    this.#queue.splice(protocolIdx, 1);
+                  }
                  this.#pendingInputQueue.splice(
                    this.#pendingInputQueue.indexOf(acceptableInput),
                    1,
@@ -316,90 +324,104 @@ export class WorkflowContext<Start = string, Stop = string, Data = unknown>
                  }
                  const [steps, inputsMap, outputsMap] =
                    this.#getStepFunction(event);
-                  const nextEventPromises: Promise<WorkflowEvent<unknown>>[] = [
-                    ...steps,
-                  ]
-                    .map((step) => {
-                      const inputs = [...(inputsMap.get(step) ?? [])];
-                      const acceptableInputs: WorkflowEvent<unknown>[] =
-                        this.#pendingInputQueue.filter((event) =>
-                          inputs.some((input) => event instanceof input),
-                        );
-                      const events: WorkflowEvent<unknown>[] = flattenEvents(
-                        inputs,
-                        [event, ...acceptableInputs],
-                      );
-                      if (events.length !== inputs.length) {
-                        if (this.#verbose) {
-                          console.log(
-                            `Not enough inputs for step ${step.name}, waiting for more events`,
+                  const nextEventPromises: Promise<WorkflowEvent<unknown> | void>[] =
+                    [...steps]
+                      .map((step) => {
+                        const inputs = [...(inputsMap.get(step) ?? [])];
+                        const acceptableInputs: WorkflowEvent<unknown>[] =
+                          this.#pendingInputQueue.filter((event) =>
+                            inputs.some((input) => event instanceof input),
                          );
-                        }
-                        // not enough to run the step, push back to the queue
-                        this.#sendEvent(event);
-                        isPendingEvents.add(event);
-                        return null;
-                      }
-                      if (isPendingEvents.has(event)) {
-                        isPendingEvents.delete(event);
-                      }
-                      if (this.#verbose) {
-                        console.log(
-                          `Running step ${step.name} with inputs ${events}`,
+                        const events: WorkflowEvent<unknown>[] = flattenEvents(
+                          inputs,
+                          [event, ...acceptableInputs],
                        );
-                      }
-                      const data = this.data;
-                      return (step as StepHandler<Data>)
-                        .call(
-                          null,
-                          {
-                            get data() {
-                              return data;
-                            },
-                            sendEvent: this.#sendEvent,
-                            requireEvent: this.#requireEvent,
-                          },
-                          // @ts-expect-error IDK why
-                          ...events.sort((a, b) => {
-                            const aIndex = inputs.indexOf(
-                              a.constructor as AnyWorkflowEventConstructor,
-                            );
-                            const bIndex = inputs.indexOf(
-                              b.constructor as AnyWorkflowEventConstructor,
-                            );
-                            return aIndex - bIndex;
-                          }),
-                        )
-                        .then((nextEvent) => {
+                        // remove the event from the queue, in case of infinite loop
+                        events.forEach((event) => {
+                          const protocolIdx = this.#queue.findIndex(
+                            (protocol) =>
+                              protocol.type === "event" &&
+                              protocol.event === event,
+                          );
+                          if (protocolIdx !== -1) {
+                            this.#queue.splice(protocolIdx, 1);
+                          }
+                        });
+                        if (events.length !== inputs.length) {
                          if (this.#verbose) {
                            console.log(
-                              `Step ${step.name} completed, next event is ${nextEvent}`,
+                              `Not enough inputs for step ${step.name}, waiting for more events`,
                            );
                          }
-                          const outputs = outputsMap.get(step) ?? [];
-                          const outputEvents = flattenEvents(outputs, [
-                            nextEvent,
-                          ]);
-                          if (outputEvents.length !== outputs.length) {
-                            if (this.#strict) {
-                              const error = Error(
-                                `Step ${step.name} returned an unexpected output event ${nextEvent}`,
+                          // not enough to run the step, push back to the queue
+                          this.#sendEvent(event);
+                          isPendingEvents.add(event);
+                          return null;
+                        }
+                        if (isPendingEvents.has(event)) {
+                          isPendingEvents.delete(event);
+                        }
+                        if (this.#verbose) {
+                          console.log(
+                            `Running step ${step.name} with inputs ${events}`,
+                          );
+                        }
+                        const data = this.data;
+                        return (step as StepHandler<Data>)
+                          .call(
+                            null,
+                            {
+                              get data() {
+                                return data;
+                              },
+                              sendEvent: this.#sendEvent,
+                              requireEvent: this.#requireEvent,
+                            },
+                            // @ts-expect-error IDK why
+                            ...events.sort((a, b) => {
+                              const aIndex = inputs.indexOf(
+                                a.constructor as AnyWorkflowEventConstructor,
                              );
-                              controller.error(error);
-                            } else {
-                              console.warn(
-                                `Step ${step.name} returned an unexpected output event ${nextEvent}`,
+                              const bIndex = inputs.indexOf(
+                                b.constructor as AnyWorkflowEventConstructor,
+                              );
+                              return aIndex - bIndex;
+                            }),
+                          )
+                          .then((nextEvent: void | WorkflowEvent<unknown>) => {
+                            if (nextEvent === undefined) {
+                              return;
+                            }
+                            if (this.#verbose) {
+                              console.log(
+                                `Step ${step.name} completed, next event is ${nextEvent}`,
                              );
                            }
-                          }
-                          if (!(nextEvent instanceof StopEvent)) {
-                            this.#pendingInputQueue.unshift(nextEvent);
-                            this.#sendEvent(nextEvent);
-                          }
-                          return nextEvent;
-                        });
-                    })
-                    .filter((promise) => promise !== null);
+                            const outputs = outputsMap.get(step) ?? [];
+                            if (
+                              !outputs.some(
+                                (output) => nextEvent.constructor === output,
+                              )
+                            ) {
+                              if (this.#strict) {
+                                const error = Error(
+                                  `Step ${step.name} returned an unexpected output event ${nextEvent}`,
+                                );
+                                controller.error(error);
+                              } else {
+                                console.warn(
+                                  `Step ${step.name} returned an unexpected output event ${nextEvent}`,
+                                );
+                              }
+                            }
+                            if (!(nextEvent instanceof StopEvent)) {
+                              this.#pendingInputQueue.unshift(nextEvent);
+                              this.#sendEvent(nextEvent);
+                            }
+                            return nextEvent;
+                          });
+                      })
+                      .filter((promise) => promise !== null);
                  nextEventPromises.forEach((promise) => {
                    pendingTasks.add(promise);
                    promise
@@ -412,6 +434,9 @@ export class WorkflowContext<Start = string, Stop = string, Data = unknown>
                  });
                  Promise.race(nextEventPromises)
                    .then((fastestNextEvent) => {
+                      if (fastestNextEvent === undefined) {
+                        return;
+                      }
                      if (!enqueuedEvents.has(fastestNextEvent)) {
                        controller.enqueue(fastestNextEvent);
                        enqueuedEvents.add(fastestNextEvent);
@@ -420,7 +445,10 @@ export class WorkflowContext<Start = string, Stop = string, Data = unknown>
                    })
                    .then(async (fastestNextEvent) =>
                      Promise.all(nextEventPromises).then((nextEvents) => {
-                        for (const nextEvent of nextEvents) {
+                        const events = nextEvents.filter(
+                          (event) => event !== undefined,
+                        );
+                        for (const nextEvent of events) {
                          // do not enqueue the same event twice
                          if (fastestNextEvent !== nextEvent) {
                            if (!enqueuedEvents.has(nextEvent)) {
@@ -432,6 +460,9 @@ export class WorkflowContext<Start = string, Stop = string, Data = unknown>
                      }),
                    )
                    .catch((err) => {
+                      // when the step raise an error, should go back to the previous step
+                      this.#sendEvent(event);
+                      isPendingEvents.add(event);
                      controller.error(err);
                    });
                }
@@ -57,10 +57,7 @@ export class Workflow<ContextData, Start, Stop> {
      AnyWorkflowEventConstructor | StartEventConstructor,
      ...(AnyWorkflowEventConstructor | StopEventConstructor)[],
    ],
-    const Out extends [
-      AnyWorkflowEventConstructor | StopEventConstructor,
-      ...(AnyWorkflowEventConstructor | StopEventConstructor)[],
-    ],
+    const Out extends (AnyWorkflowEventConstructor | StopEventConstructor)[],
  >(
    parameters: StepParameters<In, Out>,
    stepFn: (
@@ -69,9 +66,11 @@ export class Workflow<ContextData, Start, Stop> {
        [K in keyof In]: InstanceType<In[K]>;
      }
    ) => Promise<
-      {
-        [K in keyof Out]: InstanceType<Out[K]>;
-      }[number]
+      Out extends []
+        ? void
+        : {
+            [K in keyof Out]: InstanceType<Out[K]>;
+          }[number]
    >,
  ): this {
    const { inputs, outputs } = parameters;
@@ -79,6 +78,10 @@ export class Workflow<ContextData, Start, Stop> {
    return this;
  }

+  hasStep(stepFn: StepHandler): boolean {
+    return this.#steps.has(stepFn);
+  }
+
  removeStep(stepFn: StepHandler): this {
    this.#steps.delete(stepFn);
    return this;
@@ -584,13 +584,13 @@ importers:
        specifier: ^1.4.1
        version: 1.5.0
      '@llamaindex/core':
-        specifier: ^0.4.5
+        specifier: ^0.4.10
        version: link:../packages/core
      '@llamaindex/readers':
-        specifier: ^1.0.6
+        specifier: ^1.0.11
        version: link:../packages/readers
      '@llamaindex/workflow':
-        specifier: ^0.0.3
+        specifier: ^0.0.6
        version: link:../packages/workflow
      '@notionhq/client':
        specifier: ^2.2.15
@@ -617,7 +617,7 @@ importers:
        specifier: ^1.0.14
        version: 1.0.15
      llamaindex:
-        specifier: ^0.8.8
+        specifier: ^0.8.13
        version: link:../packages/llamaindex
      mongodb:
        specifier: ^6.7.0
@@ -1026,6 +1026,9 @@ importers:
      '@llamaindex/replicate':
        specifier: workspace:*
        version: link:../providers/replicate
+      '@llamaindex/vllm':
+        specifier: workspace:*
+        version: link:../providers/vllm
      '@mistralai/mistralai':
        specifier: ^1.0.4
        version: 1.1.0(zod@3.23.8)
@@ -1188,8 +1191,8 @@ importers:
  packages/providers/anthropic:
    dependencies:
      '@anthropic-ai/sdk':
-        specifier: 0.27.1
-        version: 0.27.1(encoding@0.1.13)
+        specifier: 0.32.1
+        version: 0.32.1(encoding@0.1.13)
      '@llamaindex/core':
        specifier: workspace:*
        version: link:../../core
@@ -1307,9 +1310,6 @@ importers:
      openai:
        specifier: ^4.68.1
        version: 4.69.0(encoding@0.1.13)(zod@3.23.8)
-      remeda:
-        specifier: ^2.12.0
-        version: 2.16.0
    devDependencies:
      bunchee:
        specifier: 5.6.1
@@ -1350,6 +1350,16 @@ importers:
        specifier: 5.6.1
        version: 5.6.1(typescript@5.6.3)

+  packages/providers/vllm:
+    dependencies:
+      '@llamaindex/openai':
+        specifier: workspace:*
+        version: link:../openai
+    devDependencies:
+      bunchee:
+        specifier: 5.6.1
+        version: 5.6.1(typescript@5.6.3)
+
  packages/readers:
    dependencies:
      '@azure/cosmos':
@@ -1474,6 +1484,9 @@ importers:
      tree-sitter-typescript:
        specifier: ^0.23.0
        version: 0.23.0(tree-sitter@0.22.0)
+      zod:
+        specifier: ^3.23.8
+        version: 3.23.8
    devDependencies:
      '@azure/cosmos':
        specifier: ^4.1.1
@@ -1688,6 +1701,9 @@ packages:
  '@anthropic-ai/sdk@0.27.1':
    resolution: {integrity: sha512-AKFd/E8HO26+DOVPiZpEked3Pm2feA5d4gcX2FcJXr9veDkXbKO90hr2C7N2TL7mPIMwm040ldXlsIZQ416dHg==}

+  '@anthropic-ai/sdk@0.32.1':
+    resolution: {integrity: sha512-U9JwTrDvdQ9iWuABVsMLj8nJVwAyQz6QXvgLsVhryhCEPkLsbcP/MXxm+jYcAwLoV8ESbaTTjnD4kuAFa+Hyjg==}
+
  '@apidevtools/json-schema-ref-parser@11.7.2':
    resolution: {integrity: sha512-4gY54eEGEstClvEkGnwVkTkrx0sqwemEFG5OSRRn3tD91XH0+Q8XIkYIfo7IwEWPpJZwILb9GUXeShtplRc/eA==}
    engines: {node: '>= 16'}
@@ -13412,6 +13428,18 @@ snapshots:
    transitivePeerDependencies:
      - encoding

+  '@anthropic-ai/sdk@0.32.1(encoding@0.1.13)':
+    dependencies:
+      '@types/node': 18.19.64
+      '@types/node-fetch': 2.6.11
+      abort-controller: 3.0.0
+      agentkeepalive: 4.5.0
+      form-data-encoder: 1.7.2
+      formdata-node: 4.4.1
+      node-fetch: 2.7.0(encoding@0.1.13)
+    transitivePeerDependencies:
+      - encoding
+
  '@apidevtools/json-schema-ref-parser@11.7.2':
    dependencies:
      '@jsdevtools/ono': 7.1.3
@@ -53,6 +53,9 @@
    {
      "path": "./packages/providers/ollama/tsconfig.json"
    },
+    {
+      "path": "./packages/providers/vllm/tsconfig.json"
+    },
    {
      "path": "./packages/cloud/tsconfig.json"
    },
@@ -1,5 +1,69 @@
 # @llamaindex/unit-test

+## 0.0.23
+
+### Patch Changes
+
+- ee20c44: feat(workflow): allow send event with no output
+- Updated dependencies [a7b0ac3]
+- Updated dependencies [ee20c44]
+- Updated dependencies [c69605f]
+  - @llamaindex/core@0.4.10
+  - @llamaindex/workflow@0.0.6
+  - llamaindex@0.8.13
+  - @llamaindex/cloud@2.0.10
+  - @llamaindex/node-parser@0.0.11
+  - @llamaindex/openai@0.1.35
+  - @llamaindex/readers@1.0.11
+
+## 0.0.22
+
+### Patch Changes
+
+- Updated dependencies [ea92b69]
+- Updated dependencies [fadc8b8]
+  - @llamaindex/workflow@0.0.5
+
+## 0.0.21
+
+### Patch Changes
+
+- Updated dependencies [7ae6eaa]
+  - @llamaindex/core@0.4.9
+  - @llamaindex/openai@0.1.34
+  - @llamaindex/cloud@2.0.9
+  - llamaindex@0.8.12
+  - @llamaindex/node-parser@0.0.10
+  - @llamaindex/readers@1.0.10
+
+## 0.0.20
+
+### Patch Changes
+
+- Updated dependencies [f865c98]
+  - @llamaindex/core@0.4.8
+  - @llamaindex/cloud@2.0.8
+  - llamaindex@0.8.11
+  - @llamaindex/node-parser@0.0.9
+  - @llamaindex/openai@0.1.33
+  - @llamaindex/readers@1.0.9
+
+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [f066e50]
+- Updated dependencies [d89ebe0]
+- Updated dependencies [fd8c882]
+- Updated dependencies [fd8c882]
+  - llamaindex@0.8.10
+  - @llamaindex/core@0.4.7
+  - @llamaindex/workflow@0.0.4
+  - @llamaindex/cloud@2.0.7
+  - @llamaindex/node-parser@0.0.8
+  - @llamaindex/openai@0.1.32
+  - @llamaindex/readers@1.0.8
+
 ## 0.0.18

 ### Patch Changes
@@ -1,7 +1,7 @@
 {
  "name": "@llamaindex/unit-test",
  "private": true,
-  "version": "0.0.18",
+  "version": "0.0.23",
  "type": "module",
  "scripts": {
    "test": "vitest run"
@@ -26,6 +26,7 @@
    "react-dom": "^18.3.1",
    "tree-sitter": "^0.22.0",
    "tree-sitter-javascript": "^0.23.0",
-    "tree-sitter-typescript": "^0.23.0"
+    "tree-sitter-typescript": "^0.23.0",
+    "zod": "^3.23.8"
  }
 }
@@ -0,0 +1,34 @@
+import { FunctionTool } from "@llamaindex/core/tools";
+import { describe, expect, test } from "vitest";
+import { z } from "zod";
+
+describe("function-tool", () => {
+  test("zod type check", () => {
+    const inputSchema = z.object({
+      name: z.string(),
+      age: z.coerce.number(),
+    });
+    const tool = FunctionTool.from(
+      (input) => {
+        if (typeof input.age !== "number") {
+          throw new Error("Age should be a number");
+        }
+        return "Hello " + input.name + " " + input.age;
+      },
+      {
+        name: "get-user",
+        description: "Get user by name and age",
+        parameters: inputSchema,
+      },
+    );
+    {
+      const response = tool.call({ name: "John", age: 30 });
+      expect(response).toBe("Hello John 30");
+    }
+    {
+      // @ts-expect-error age should be a number
+      const response = tool.call({ name: "John", age: "30" });
+      expect(response).toBe("Hello John 30");
+    }
+  });
+});
@@ -531,6 +531,21 @@ describe("workflow basic", () => {
    const result = await myWorkflow.run("start");
    expect(result.data).toBe("query result");
  });
+
+  test("allow output with send event", async () => {
+    const myFlow = new Workflow<unknown, string, string>({ verbose: true });
+    myFlow.addStep(
+      {
+        inputs: [StartEvent<string>],
+        outputs: [],
+      },
+      async (context, ev) => {
+        context.sendEvent(new StopEvent(`Hello ${ev.data}!`));
+      },
+    );
+    const result = myFlow.run("world");
+    expect((await result).data).toBe("Hello world!");
+  });
 });

 describe("workflow event loop", () => {
@@ -794,6 +809,21 @@ describe("workflow event loop", () => {
      }
    `);
  });
+
+  test("workflow multiple output", async () => {
+    const myFlow = new Workflow<unknown, string, string>({ verbose: true });
+    myFlow.addStep(
+      {
+        inputs: [StartEvent<string>],
+        outputs: [StopEvent<string>, StopEvent<string>],
+      },
+      async (_context, ev) => {
+        return new StopEvent(`Hello ${ev.data}!`);
+      },
+    );
+    const result = await myFlow.run("world").strict();
+    expect(result.data).toBe("Hello world!");
+  });
 });

 describe("snapshot", async () => {
@@ -869,3 +899,83 @@ describe("snapshot", async () => {
    expect(fn).toHaveBeenCalledTimes(1);
  });
 });
+
+describe("error", () => {
+  test("error in handler", async () => {
+    const myFlow = new Workflow<boolean, string, string>({ verbose: true });
+    myFlow.addStep(
+      {
+        inputs: [StartEvent<string>],
+        outputs: [StopEvent<string>],
+      },
+      async ({ data }) => {
+        if (!data) {
+          throw new Error("Something went wrong");
+        } else {
+          return new StopEvent(`Hello ${data}!`);
+        }
+      },
+    );
+    await expect(myFlow.run("world")).rejects.toThrow("Something went wrong");
+    {
+      const context = myFlow.run("world");
+      try {
+        for await (const _ of context) {
+          // do nothing
+        }
+      } catch (error) {
+        expect(error).toBeInstanceOf(Error);
+        expect((error as Error).message).toBe("Something went wrong");
+        const snapshot = context.snapshot();
+        const newContext = myFlow.recover(snapshot).with(true);
+        expect((await newContext).data).toBe("Hello true!");
+      }
+    }
+  });
+
+  test("recover in the middle of workflow", async () => {
+    const myFlow = new Workflow<string | undefined, string, string>({
+      verbose: true,
+    });
+
+    class AEvent extends WorkflowEvent<string> {}
+
+    myFlow.addStep(
+      {
+        inputs: [StartEvent<string>],
+        outputs: [AEvent],
+      },
+      async ({ data }) => {
+        if (data !== undefined) {
+          throw new Error("Something went wrong");
+        }
+        return new AEvent("world");
+      },
+    );
+    myFlow.addStep(
+      {
+        inputs: [AEvent],
+        outputs: [StopEvent],
+      },
+      async ({ data }, ev) => {
+        if (data === undefined) {
+          throw new Error("Something went wrong");
+        }
+        return new StopEvent(`Hello, ${data}!`);
+      },
+    );
+    // no context, so will throw error
+    const context = myFlow.run("world");
+    try {
+      for await (const _ of context) {
+        // do nothing
+      }
+    } catch (error) {
+      expect(error).toBeInstanceOf(Error);
+      expect((error as Error).message).toBe("Something went wrong");
+      const snapshot = context.snapshot();
+      const newContext = myFlow.recover(snapshot).with("Recovered Data");
+      expect((await newContext).data).toBe("Hello, Recovered Data!");
+    }
+  });
+});
Author	SHA1	Message	Date
github-actions[bot]	4c7b891446	Release 0.8.13 (#1480 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: himself65 <himself65@users.noreply.github.com>	2024-11-14 18:25:19 -08:00
Alex Yang	a9c5b4899b	docs: update docusaurus.config.js (#1489 )	2024-11-14 17:36:52 -08:00
Alex Yang	a7b0ac3cb7	feat(anthropic): support prompt caching (#1488 )	2024-11-14 17:35:21 -08:00
Alex Yang	a7540ff47b	docs: update (#1486 )	2024-11-14 15:46:48 -08:00
Aman Rao	c69605f406	feat: add async support to BaseChatStore and BaseChatStoreMemory (#1483 ) Co-authored-by: Alex Yang <himself65@outlook.com>	2024-11-14 13:45:50 -08:00
Alex Yang	ee20c44d9b	feat(workflow): allow send event with no output (#1479 )	2024-11-14 00:17:53 -08:00
Alex Yang	1d470363df	docs: add banner for legacy (#1478 )	2024-11-13 18:16:13 -08:00
github-actions[bot]	b39f40dbd8	Release (#1477 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: himself65 <himself65@users.noreply.github.com>	2024-11-13 01:25:40 -08:00
Alex Yang	fadc8b8ea0	feat: recoverable data with error handling (#1476 )	2024-11-13 01:15:50 -08:00
Alex Yang	ea92b6986d	chore: update changeset	2024-11-13 01:15:28 -08:00
Alex Yang	17f9022d22	fix: output event check (#1475 )	2024-11-13 00:46:35 -08:00
github-actions[bot]	14792cd8b4	Release 0.8.12 (#1473 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>	2024-11-12 16:20:25 -08:00
Alex Yang	7ae6eaa0a2	chore: update changeset	2024-11-12 12:49:17 -08:00
Alex Yang	dbb5bd9f23	feat: allow `tool_choice` for OpenAIAgent (#1472 )	2024-11-12 12:46:57 -08:00
github-actions[bot]	aacd606204	Release 0.8.11 (#1471 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>	2024-11-12 11:49:22 -08:00
Alex Yang	f865c984d3	feat: async get message on chat store (#1470 )	2024-11-12 10:59:44 -08:00
github-actions[bot]	7b10882d06	Release 0.8.10 (#1466 ) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: himself65 <himself65@users.noreply.github.com>	2024-11-11 14:19:46 -08:00
Alex Yang	f066e50482	feat: vllm support (#1468 )	2024-11-11 13:14:08 -08:00
Alex Yang	fd8c882792	refactor: migrate example to new workflow API (#1467 )	2024-11-11 12:03:38 -08:00
Alex Yang	d89ebe0261	chore: update changeset	2024-11-11 10:11:04 -08:00
Alex Yang	968feb32cd	feat: better input type for function tool with `zod` (#1464 )	2024-11-11 10:10:03 -08:00
Alex Yang	43f6f56c5b	docs(next): fix turbo.json (#1465 )	2024-11-11 10:07:12 -08:00
				`@@ -0,0 +1 @@`
				`export { VLLM, type VLLMParams } from "./llm";`