Merge branch 'main' into feat/build-wasm-with-extism

.changeset/quick-brooms-sniff.md

@@ -0,0 +1,5 @@
+---
+"llamaindex": patch
+---
+
+feat: update JinaAIEmbedding, support embedding v3

.changeset/shy-bottles-shop.md

@@ -0,0 +1,5 @@
+---
+"@llamaindex/extism-tools": patch
+---
+
+Build WASM tools with extism

.github/workflows/lint_on_push_or_pull.yml

@@ -19,6 +19,10 @@ jobs:
         with:
           node-version-file: ".nvmrc"
           cache: "pnpm"
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
       - name: Install dependencies
         run: pnpm install
       - name: Run lint

.github/workflows/preview.yml

@@ -18,6 +18,11 @@ jobs:
           node-version-file: ".nvmrc"
           cache: "pnpm"
 
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
+
       - name: Install dependencies
         run: pnpm install
 

.github/workflows/release.yml

@@ -23,6 +23,11 @@ jobs:
           node-version-file: ".nvmrc"
           cache: "pnpm"
 
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
+
       - name: Install dependencies
         run: pnpm install
 

.github/workflows/test.yml

@@ -42,6 +42,10 @@ jobs:
         with:
           node-version: ${{ matrix.node-version }}
           cache: "pnpm"
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
       - name: Install dependencies
         run: pnpm install
       - name: Run E2E Tests
@@ -62,6 +66,10 @@ jobs:
         with:
           node-version: ${{ matrix.node-version }}
           cache: "pnpm"
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
       - name: Install dependencies
         run: pnpm install
       - name: Run tests
@@ -77,6 +85,10 @@ jobs:
         with:
           node-version-file: ".nvmrc"
           cache: "pnpm"
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
       - name: Install dependencies
         run: pnpm install
       - name: Build
@@ -114,6 +126,10 @@ jobs:
         with:
           node-version-file: ".nvmrc"
           cache: "pnpm"
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
       - name: Install dependencies
         run: pnpm install
       - name: Build llamaindex
@@ -133,6 +149,10 @@ jobs:
         with:
           node-version-file: ".nvmrc"
           cache: "pnpm"
+      - name: Install Extism
+        run: |
+          curl -O https://raw.githubusercontent.com/extism/js-pdk/main/install.sh
+          sh install.sh
       - name: Install dependencies
         run: pnpm install
       - name: Build

apps/docs/docs/modules/agent/_category_.yml

@@ -1,2 +1,2 @@
 label: "Agents"
-position: 3
+position: 10

apps/docs/docs/modules/chat_engine.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 4
+sidebar_position: 13
 ---
 
 # ChatEngine

apps/docs/docs/modules/data_index.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 4
+sidebar_position: 12
 ---
 
 # Index
@@ -8,6 +8,7 @@ An index is the basic container and organization for your data. LlamaIndex.TS su
 
 - `VectorStoreIndex` - will send the top-k `Node`s to the LLM when generating a response. The default top-k is 2.
 - `SummaryIndex` - will send every `Node` in the index to the LLM in order to generate a response
+- `KeywordTableIndex` extracts and provides keywords from `Node`s to the LLM
 
 ```typescript
 import { Document, VectorStoreIndex } from "llamaindex";

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

@@ -6,6 +6,19 @@ import CodeSource2 from "!raw-loader!../../../../../examples/readers/src/custom-
 
 Before you can start indexing your documents, you need to load them into memory.
 
+All "basic" data loaders can be seen below, mapped to their respective filetypes in `SimpleDirectoryReader`. More loaders are shown in the sidebar on the left.
+Additionally the following loaders exist without separate documentation:
+
+- `AssemblyAIReader` transcribes audio using [AssemblyAI](https://www.assemblyai.com/).
+  - [AudioTranscriptReader](../../api/classes/AudioTranscriptReader.md): loads entire transcript as a single document.
+  - [AudioTranscriptParagraphsReader](../../api/classes/AudioTranscriptParagraphsReader.md): creates a document per paragraph.
+  - [AudioTranscriptSentencesReader](../../api/classes/AudioTranscriptSentencesReader.md): creates a document per sentence.
+  - [AudioSubtitlesReader](../../api/classes/AudioTranscriptParagraphsReader.md): creates a document containing the subtitles of a transcript.
+- [NotionReader](../../api/classes/NotionReader.md) loads [Notion](https://www.notion.so/) pages.
+- [SimpleMongoReader](../../api/classes/SimpleMongoReader) loads data from a [MongoDB](https://www.mongodb.com/).
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
 ## SimpleDirectoryReader
 
 [![Open in StackBlitz](https://developer.stackblitz.com/img/open_in_stackblitz.svg)](https://stackblitz.com/github/run-llama/LlamaIndexTS/tree/main/examples/readers?file=src/simple-directory-reader.ts&title=Simple%20Directory%20Reader)

apps/docs/docs/modules/data_stores/_category_.yml

@@ -0,0 +1,2 @@
+label: "Data Stores"
+position: 2

apps/docs/docs/modules/data_stores/chat_stores/_category_.yml

@@ -0,0 +1 @@
+label: "Chat Stores"

apps/docs/docs/modules/data_stores/chat_stores/index.md

@@ -0,0 +1,13 @@
+# Chat Stores
+
+Chat stores manage chat history by storing sequences of messages in a structured way, ensuring the order of messages is maintained for accurate conversation flow.
+
+## Available Chat Stores
+
+- [SimpleChatStore](../../../api/classes/SimpleChatStore.md): A simple in-memory chat store with support for [persisting](../index.md#local-storage) data to disk.
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
+## API Reference
+
+- [BaseChatStore](../../../api/interfaces/BaseChatStore.md)

apps/docs/docs/modules/data_stores/doc_stores/_category_.yml

@@ -0,0 +1,2 @@
+label: "Document Stores"
+position: 2

apps/docs/docs/modules/data_stores/doc_stores/index.md

@@ -0,0 +1,14 @@
+# Document Stores
+
+Document stores contain ingested document chunks, i.e. [Node](../../documents_and_nodes/index.md)s.
+
+## Available Document Stores
+
+- [SimpleDocumentStore](../../../api/classes/SimpleDocumentStore.md): A simple in-memory document store with support for [persisting](../index.md#local-storage) data to disk.
+- [PostgresDocumentStore](../../../api/classes/PostgresDocumentStore.md): A PostgreSQL document store, see [PostgreSQL Storage](../index.md#postgresql-storage).
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
+## API Reference
+
+- [BaseDocumentStore](../../../api/classes/BaseDocumentStore.md)

apps/docs/docs/modules/data_stores/index.md

@@ -1,7 +1,3 @@
----
-sidebar_position: 7
----
-
 # Storage
 
 Storage in LlamaIndex.TS works automatically once you've configured a
@@ -57,4 +53,4 @@ const index = await VectorStoreIndex.fromDocuments([document], {
 
 ## API Reference
 
-- [StorageContext](../api/interfaces/StorageContext.md)
+- [StorageContext](../../api/interfaces/StorageContext.md)

apps/docs/docs/modules/data_stores/index_stores/_category_.yml

@@ -0,0 +1,2 @@
+label: "Index Stores"
+position: 3

apps/docs/docs/modules/data_stores/index_stores/index.md

@@ -0,0 +1,14 @@
+# Index Stores
+
+Index stores are underlying storage components that contain metadata(i.e. information created when indexing) about the [index](../../data_index.md) itself.
+
+## Available Index Stores
+
+- [SimpleIndexStore](../../../api/classes/SimpleIndexStore.md): A simple in-memory index store with support for [persisting](../index.md#local-storage) data to disk.
+- [PostgresIndexStore](../../../api/classes/PostgresIndexStore.md): A PostgreSQL index store, , see [PostgreSQL Storage](../index.md#postgresql-storage).
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
+## API Reference
+
+- [BaseIndexStore](../../../api/classes/BaseIndexStore.md)

apps/docs/docs/modules/data_stores/kv_stores/_category_.yml

@@ -0,0 +1,2 @@
+label: "Key-Value Stores"
+position: 4

apps/docs/docs/modules/data_stores/kv_stores/index.md

@@ -0,0 +1,14 @@
+# Key-Value Stores
+
+Key-Value Stores represent underlying storage components used in [Document Stores](../doc_stores/index.md) and [Index Stores](../index_stores/index.md)
+
+## Available Key-Value Stores
+
+- [SimpleKVStore](../../../api/classes/SimpleKVStore.md): A simple Key-Value store with support of [persisting](../index.md#local-storage) data to disk.
+- [PostgresKVStore](../../../api/classes/PostgresKVStore.md): A PostgreSQL Key-Value store, see [PostgreSQL Storage](../index.md#postgresql-storage).
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
+## API Reference
+
+- [BaseKVStore](../../../api/classes/BaseKVStore.md)

apps/docs/docs/modules/data_stores/vector_stores/index.md

@@ -0,0 +1,22 @@
+# Vector Stores
+
+Vector stores save embedding vectors of your ingested document chunks.
+
+## Available Vector Stores
+
+Available Vector Stores are shown on the sidebar to the left. Additionally the following integrations exist without separate documentation:
+
+- [SimpleVectorStore](../../../api/classes/SimpleVectorStore.md): A simple in-memory vector store with optional [persistance](../index.md#local-storage) to disk.
+- [AstraDBVectorStore](../../../api/classes/AstraDBVectorStore.md): A cloud-native, scalable Database-as-a-Service built on Apache Cassandra, see [datastax.com](https://www.datastax.com/products/datastax-astra)
+- [ChromaVectorStore](../../../api/classes/ChromaVectorStore.md): An open-source vector database, focused on ease of use and performance, see [trychroma.com](https://www.trychroma.com/)
+- [MilvusVectorStore](../../../api/classes/MilvusVectorStore.md): An open-source, high-performance, highly scalable vector database, see [milvus.io](https://milvus.io/)
+- [MongoDBAtlasVectorSearch](../../../api/classes/MongoDBAtlasVectorSearch.md): A cloud-based vector search solution for MongoDB, see [mongodb.com](https://www.mongodb.com/products/platform/atlas-vector-search)
+- [PGVectorStore](../../../api/classes/PGVectorStore.md): An open-source vector store built on PostgreSQL, see [pgvector Github](https://github.com/pgvector/pgvector)
+- [PineconeVectorStore](../../../api/classes/PineconeVectorStore.md): A managed, cloud-native vector database, see [pinecone.io](https://www.pinecone.io/)
+- [WeaviateVectorStore](../../../api/classes/WeaviateVectorStore.md): An open-source, ai-native vector database, see [weaviate.io](https://weaviate.io/)
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
+## API Reference
+
+- [VectorStoreBase](../../../api/classes/VectorStoreBase.md)

apps/docs/docs/modules/data_stores/vector_stores/qdrant.md

@@ -1,5 +1,7 @@
 # Qdrant Vector Store
 
+[qdrant.tech](https://qdrant.tech/)
+
 To run this example, you need to have a Qdrant instance running. You can run it with Docker:
 
 ```bash
@@ -87,4 +89,4 @@ main().catch(console.error);
 
 ## API Reference
 
-- [QdrantVectorStore](../../api/classes/QdrantVectorStore.md)
+- [QdrantVectorStore](../../../api/classes/QdrantVectorStore.md)

apps/docs/docs/modules/documents_and_nodes/index.md

@@ -1,7 +1,3 @@
----
-sidebar_position: 1
----
-
 # Documents and Nodes
 
 `Document`s and `Node`s are the basic building blocks of any index. While the API for these objects is similar, `Document` objects represent entire files, while `Node`s are smaller pieces of that original document, that are suitable for an LLM and Q&A.

apps/docs/docs/modules/embeddings/_category_.yml

@@ -1,2 +1,2 @@
 label: "Embeddings"
-position: 3
+position: 6

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

@@ -7,7 +7,7 @@ To find out more about the latest features, updates, and available models, visit
 ## Table of Contents
 
 1. [Setup](#setup)
-2. [Usage with LlamaIndex](#integration-with-llamaindex)
+2. [Usage with LlamaIndex](#usage-with-llamaindex)
 3. [Embeddings with Custom Parameters](#embeddings-with-custom-parameters)
 
 ## Setup

apps/docs/docs/modules/embeddings/index.md

@@ -16,6 +16,16 @@ Settings.embedModel = new OpenAIEmbedding({
 
 For local embeddings, you can use the [HuggingFace](./available_embeddings/huggingface.md) embedding model.
 
+## Available Embeddings
+
+Most available embeddings are listed in the sidebar on the left.
+Additionally the following integrations exist without separate documentation:
+
+- [ClipEmbedding](../../api/classes/ClipEmbedding.md) using `@xenova/transformers`
+- [FireworksEmbedding](../../api/classes/FireworksEmbedding.md) see [fireworks.ai](https://fireworks.ai/)
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
 ## API Reference
 
 - [OpenAIEmbedding](../../api/classes/OpenAIEmbedding.md)

apps/docs/docs/modules/evaluation/_category_.yml

@@ -1,2 +1,2 @@
 label: "Evaluating"
-position: 3
+position: 9

apps/docs/docs/modules/ingestion_pipeline/_category_.yml

@@ -1,2 +1,2 @@
 label: "Ingestion Pipeline"
-position: 2
+position: 4

apps/docs/docs/modules/llms/_category_.yml

@@ -1,2 +1,2 @@
 label: "LLMs"
-position: 3
+position: 5

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

@@ -1,6 +1,6 @@
 # Fireworks LLM
 
-Fireworks.ai focus on production use cases for open source LLMs, offering speed and quality.
+[Fireworks.ai](https://fireworks.ai/) focus on production use cases for open source LLMs, offering speed and quality.
 
 ## Usage
 

apps/docs/docs/modules/llms/index.md

@@ -1,7 +1,3 @@
----
-sidebar_position: 3
----
-
 # Large Language Models (LLMs)
 
 The LLM is responsible for reading text and generating natural language responses to queries. By default, LlamaIndex.TS uses `gpt-3.5-turbo`.
@@ -30,6 +26,15 @@ export AZURE_OPENAI_DEPLOYMENT="gpt-4" # or some other deployment name
 
 For local LLMs, currently we recommend the use of [Ollama](./available_llms/ollama.md) LLM.
 
+## Available LLMs
+
+Most available LLMs are listed in the sidebar on the left. Additionally the following integrations exist without separate documentation:
+
+- [HuggingFaceLLM](../../api/classes/HuggingFaceLLM.md) and [HuggingFaceInferenceAPI](../../api/classes/HuggingFaceInferenceAPI.md).
+- [ReplicateLLM](../../api/classes/ReplicateLLM.md) see [replicate.com](https://replicate.com/)
+
+Check the [LlamaIndexTS Github](https://github.com/run-llama/LlamaIndexTS) for the most up to date overview of integrations.
+
 ## API Reference
 
 - [OpenAI](../../api/classes/OpenAI.md)

apps/docs/docs/modules/node_parser.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 4
+sidebar_position: 11
 ---
 
 # NodeParser

apps/docs/docs/modules/node_postprocessors/index.md

@@ -107,3 +107,4 @@ const filteredNodes = processor.postprocessNodes(nodes);
 ## API Reference
 
 - [SimilarityPostprocessor](../../api/classes/SimilarityPostprocessor.md)
+- [MetadataReplacementPostProcessor](../../api/classes/MetadataReplacementPostProcessor.md)

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

@@ -7,7 +7,7 @@ To find out more about the latest features and updates, visit the [mixedbread.ai
 ## Table of Contents
 
 1. [Setup](#setup)
-2. [Usage with LlamaIndex](#integration-with-llamaindex)
+2. [Usage with LlamaIndex](#usage-with-llamaindex)
 3. [Simple Reranking Guide](#simple-reranking-guide)
 4. [Reranking with Objects](#reranking-with-objects)
 

apps/docs/docs/modules/prompt/_category_.yml

@@ -1,2 +1,2 @@
 label: "Prompts"
-position: 0
+position: 7

apps/docs/docs/modules/prompt/index.md

@@ -73,6 +73,5 @@ const response = await queryEngine.query({
 
 ## API Reference
 
-- [TextQaPrompt](../../api/type-aliases/TextQaPrompt.md)
 - [ResponseSynthesizer](../../api/classes/ResponseSynthesizer.md)
 - [CompactAndRefine](../../api/classes/CompactAndRefine.md)

apps/docs/docs/modules/query_engines/_category_.yml

@@ -1,2 +1,2 @@
 label: "Query Engines"
-position: 2
+position: 8

apps/docs/docs/modules/response_synthesizer.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 6
+sidebar_position: 15
 ---
 
 # ResponseSynthesizer

apps/docs/docs/modules/retriever.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 5
+sidebar_position: 14
 ---
 
 # Retriever

apps/docs/package.json

@@ -37,7 +37,7 @@
     "docusaurus-plugin-typedoc": "1.0.5",
     "typedoc": "0.26.6",
     "typedoc-plugin-markdown": "4.2.6",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   },
   "browserslist": {
     "production": [

examples/package.json

@@ -21,7 +21,7 @@
   "devDependencies": {
     "@types/node": "^22.5.1",
     "tsx": "^4.19.0",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   },
   "scripts": {
     "lint": "eslint ."

examples/readers/package.json

@@ -23,6 +23,6 @@
   "devDependencies": {
     "@types/node": "^22.5.1",
     "tsx": "^4.19.0",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   }
 }

package.json

@@ -2,9 +2,9 @@
   "name": "@llamaindex/monorepo",
   "private": true,
   "scripts": {
-    "build": "turbo run build --filter=\"!docs\" --filter=\"!*-test\" --filter=\"!*-example\"",
-    "build:release": "turbo run build lint test --filter=\"!docs\" --filter=\"!*-test\" --filter=\"!*-example\"",
-    "dev": "turbo run dev",
+    "build": "turbo run build",
+    "build:release": "turbo run build --filter=\"./packages/*\"",
+    "dev": "turbo run dev --filter=\"./packages/*\"",
     "format": "prettier --ignore-unknown --cache --check .",
     "format:write": "prettier --ignore-unknown --write .",
     "lint": "turbo run lint",
@@ -31,7 +31,7 @@
     "prettier": "^3.3.3",
     "prettier-plugin-organize-imports": "^4.0.0",
     "turbo": "^2.1.0",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   },
   "packageManager": "pnpm@9.5.0",
   "pnpm": {

packages/autotool/examples/02_nextjs/actions.ts

@@ -5,9 +5,9 @@ import { runWithStreamableUI } from "@/context";
 import "@/tool";
 import { convertTools } from "@llamaindex/autotool";
 import { createStreamableUI } from "ai/rsc";
-import type { JSX } from "react";
+import type { ReactNode } from "react";
 
-export async function chatWithAI(message: string): Promise<JSX.Element> {
+export async function chatWithAI(message: string): Promise<ReactNode> {
   const agent = new OpenAIAgent({
     tools: convertTools("llamaindex"),
   });
@@ -25,7 +25,7 @@ export async function chatWithAI(message: string): Promise<JSX.Element> {
               uiStream.append("\n");
             },
             write: async (message) => {
-              uiStream.append(message.response.delta);
+              uiStream.append(message.response);
             },
             close: () => {
               uiStream.done();

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

@@ -32,6 +32,6 @@
     "cross-env": "^7.0.3",
     "postcss": "^8.4.41",
     "tailwindcss": "^3.4.10",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   }
 }

packages/autotool/package.json

@@ -51,7 +51,7 @@
     "unplugin": "^1.12.2"
   },
   "peerDependencies": {
-    "llamaindex": "^0.6.0",
+    "llamaindex": "workspace:*",
     "openai": "^4",
     "typescript": "^4"
   },
@@ -75,7 +75,7 @@
     "next": "14.2.7",
     "rollup": "^4.21.2",
     "tsx": "^4.19.0",
-    "typescript": "^5.5.4",
+    "typescript": "^5.6.2",
     "vitest": "^2.0.5",
     "webpack": "^5.94.0"
   }

packages/autotool/src/compiler.ts

@@ -9,7 +9,7 @@ import td from "typedoc";
 import type { SourceMapCompact } from "unplugin";
 import type { InfoString } from "./internal";
 
-export const isToolFile = (url: string) => /tool\.[jt]sx?$/.test(url);
+export const isToolFile = (url: string) => /\.tool\.[jt]sx?$/.test(url);
 export const isJSorTS = (url: string) => /\.m?[jt]sx?$/.test(url);
 
 async function parseRoot(entryPoint: string) {
@@ -28,7 +28,7 @@ async function parseRoot(entryPoint: string) {
   if (project) {
     return app.serializer.projectToObject(project, process.cwd());
   }
-  throw new Error("Failed to parse root");
+  throw new Error(`Failed to parse root ${entryPoint}`);
 }
 
 export async function transformAutoTool(

packages/core/package.json

@@ -172,8 +172,10 @@
     "url": "https://github.com/himself65/LlamaIndexTS.git"
   },
   "devDependencies": {
+    "@edge-runtime/vm": "^4.0.3",
     "ajv": "^8.17.1",
     "bunchee": "5.3.2",
+    "happy-dom": "^15.7.4",
     "natural": "^8.0.1",
     "python-format-js": "^1.4.3"
   },

packages/core/src/workflow/context.ts

@@ -95,7 +95,7 @@ export class Context {
     this.#streamingQueue.push(event);
   }
 
-  async *streamEvents(): AsyncGenerator<WorkflowEvent, void, undefined> {
+  async *streamEvents(): AsyncGenerator<WorkflowEvent, void, void> {
     while (true) {
       const event = this.#streamingQueue.shift();
       if (event) {

packages/core/src/workflow/workflow.ts

@@ -62,7 +62,7 @@ export class Workflow {
     return stepInfo.inputs.includes(eventType);
   }
 
-  async *streamEvents(): AsyncGenerator<WorkflowEvent, void, unknown> {
+  async *streamEvents(): AsyncGenerator<WorkflowEvent, void> {
     if (this.#contexts.size > 1) {
       throw new Error(
         "This workflow has multiple concurrent runs in progress and cannot stream events. " +

packages/core/tests/vitest.workspace.ts

@@ -0,0 +1,19 @@
+import { defineWorkspace } from "vitest/config";
+
+export default defineWorkspace([
+  {
+    test: {
+      environment: "edge-runtime",
+    },
+  },
+  {
+    test: {
+      environment: "happy-dom",
+    },
+  },
+  {
+    test: {
+      environment: "node",
+    },
+  },
+]);

packages/extism-tools/README.md

@@ -0,0 +1,24 @@
+## Extism Tools
+
+### Prerequisites for Development
+
+- [Extism PDK](https://github.com/extism/js-pdk?tab=readme-ov-file#linux-macos)
+
+### Build WASM files
+
+```bash
+pnpm run build
+```
+
+### Run WASM files in Node.js using Extism SDK (https://github.com/extism/js-sdk)
+
+```bash
+cd examples
+pnpm run test:wiki
+```
+
+### Run WASM files in Python using Extism SDK (https://github.com/extism/python-sdk)
+
+```bash
+python examples/wasm/wiki.py
+```

packages/extism-tools/bin/compile.js

@@ -0,0 +1,31 @@
+// needed as extism-js doesn't support compiling multiple files
+import { execSync } from "child_process";
+import { mkdirSync, readdirSync } from "fs";
+
+const WASM_SRC_FOLDER = "wasm";
+const WASM_OUTPUT_FOLDER = "dist/wasm";
+
+// get list of tools from files (except index.d.ts)
+const tools = readdirSync(WASM_SRC_FOLDER)
+  .filter((file) => !file.includes("index.d.ts"))
+  .map((file) => file.split(".")[0]);
+
+// create dist/wasm folder if it doesn't exist using fs
+try {
+  mkdirSync(WASM_OUTPUT_FOLDER, { recursive: true });
+} catch (error) {
+  console.error("Error creating dist/wasm folder:", error.message);
+  process.exit(1);
+}
+
+// loop through each tool, compile it to wasm using extism-js
+tools.forEach((tool) => {
+  try {
+    execSync(
+      `extism-js ${WASM_SRC_FOLDER}/${tool}.js -i ${WASM_SRC_FOLDER}/index.d.ts -o ${WASM_OUTPUT_FOLDER}/${tool}.wasm`,
+    );
+  } catch (error) {
+    console.error(`Error compiling module ${tool}:`, error.message);
+    process.exit(1);
+  }
+});

packages/extism-tools/examples/package.json

@@ -0,0 +1,17 @@
+{
+  "name": "@llamaindex/extism-tools-examples",
+  "private": true,
+  "version": "1.0.0",
+  "type": "module",
+  "scripts": {
+    "test:wiki": "tsx ./src/wiki.ts",
+    "test:todo": "tsx ./src/todo.ts"
+  },
+  "dependencies": {
+    "@llamaindex/extism-tools": "workspace:*",
+    "llamaindex": "workspace:*"
+  },
+  "devDependencies": {
+    "tsx": "^4.19.0"
+  }
+}

packages/extism-tools/examples/src/todo.ts

@@ -0,0 +1,13 @@
+import { ExtismToolFactory } from "@llamaindex/extism-tools";
+import { OpenAI, OpenAIAgent, Settings } from "llamaindex";
+
+async function main() {
+  const TodoTool = await ExtismToolFactory.createToolClass("todo");
+  const todoTool = new TodoTool();
+  Settings.llm = new OpenAI();
+  const agent = new OpenAIAgent({ tools: [todoTool] });
+  const result = await agent.chat({ message: "Get first todo" });
+  console.log(result.message);
+}
+
+void main();

packages/extism-tools/examples/src/wiki.py

@@ -0,0 +1,34 @@
+import extism
+import json
+from os.path import join, dirname
+
+
+def read_local_wasm(file_name):
+    path = join(dirname(__file__), file_name)  # Change this to your wasm file path
+    with open(path, "rb") as wasm_file:
+        return wasm_file.read()
+
+
+def _manifest(file_name):
+    wasm = read_local_wasm(file_name)
+    return {
+        "wasm": [{"data": wasm}],
+        "allowed_hosts": ["*.wikipedia.org"],
+    }
+
+
+manifest = _manifest("wiki.wasm")
+with extism.Plugin(manifest, wasi=True) as plugin:
+    metadata = plugin.call(
+        "getMetadata",
+        "",
+        parse=lambda output: json.loads(bytes(output).decode("utf-8")),
+    )
+    data = plugin.call(
+        "call",
+        json.dumps({"query": "Ho Chi Minh City"}),
+        parse=lambda output: json.loads(bytes(output).decode("utf-8")),
+    )
+
+print(metadata)
+print(data)

packages/extism-tools/examples/src/wiki.ts

@@ -0,0 +1,13 @@
+import { ExtismToolFactory } from "@llamaindex/extism-tools";
+import { OpenAI, OpenAIAgent, Settings } from "llamaindex";
+
+async function main() {
+  const WikiTool = await ExtismToolFactory.createToolClass("wiki");
+  const wikiTool = new WikiTool();
+  Settings.llm = new OpenAI();
+  const agent = new OpenAIAgent({ tools: [wikiTool] });
+  const result = await agent.chat({ message: "Ho Chi Minh City" });
+  console.log(result.message);
+}
+
+void main();

packages/extism-tools/examples/tsconfig.json

@@ -0,0 +1,9 @@
+{
+  "extends": "../../../tsconfig.json",
+  "compilerOptions": {
+    "outDir": "./lib",
+    "module": "node16",
+    "moduleResolution": "node16"
+  },
+  "include": ["./src"]
+}

packages/extism-tools/package.json

@@ -0,0 +1,62 @@
+{
+  "name": "@llamaindex/extism-tools",
+  "version": "0.0.1",
+  "license": "MIT",
+  "type": "module",
+  "dependencies": {
+    "@extism/extism": "^2.0.0-rc8",
+    "ajv": "^8.17.1",
+    "@llamaindex/core": "workspace:*",
+    "@llamaindex/env": "workspace:*",
+    "llamaindex": "workspace:*"
+  },
+  "devDependencies": {
+    "@swc/cli": "^0.4.0",
+    "@swc/core": "^1.7.22",
+    "typescript": "^5.5.4",
+    "@types/node": "^22.5.1"
+  },
+  "engines": {
+    "node": ">=18.0.0"
+  },
+  "types": "./dist/index.d.ts",
+  "main": "./dist/cjs/index.js",
+  "exports": {
+    ".": {
+      "import": {
+        "types": "./dist/index.d.ts",
+        "default": "./dist/index.js"
+      },
+      "require": {
+        "types": "./dist/index.d.ts",
+        "default": "./dist/cjs/index.js"
+      }
+    },
+    "./*": {
+      "import": {
+        "types": "./dist/*.d.ts",
+        "default": "./dist/*.js"
+      },
+      "require": {
+        "types": "./dist/*.d.ts",
+        "default": "./dist/cjs/*.js"
+      }
+    }
+  },
+  "files": [
+    "dist",
+    "CHANGELOG.md"
+  ],
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/run-llama/LlamaIndexTS.git",
+    "directory": "packages/extism-tools"
+  },
+  "scripts": {
+    "build": "rm -rf ./dist && pnpm run build:wasm && pnpm run build:esm && pnpm run build:cjs && pnpm run build:type",
+    "build:esm": "swc src -d dist --strip-leading-paths --config-file ../../.swcrc",
+    "build:cjs": "swc src -d dist/cjs --strip-leading-paths --config-file ../../.cjs.swcrc",
+    "build:type": "tsc -p tsconfig.json",
+    "build:wasm": "node bin/compile.js"
+  }
+}

packages/extism-tools/src/ExtismToolFactory.ts

@@ -0,0 +1,82 @@
+import createPlugin, { type Plugin } from "@extism/extism";
+import type { JSONSchemaType } from "ajv";
+import type { BaseToolWithCall, ToolMetadata } from "llamaindex";
+import path from "path";
+import { fileURLToPath } from "url";
+
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+export const WASM_DIRECTORY = path.join(__dirname, "..", "dist", "wasm");
+export const DEFAULT_MAX_HTTP_RESPONSE_BYTES = 100 * 1024 * 1024; // 100 MB
+
+export type ToolParams = Record<string, any>;
+
+export type ToolClassParams = {
+  metadata?: ToolMetadata<JSONSchemaType<ToolParams>>;
+};
+
+export type CreateToolClassParams = {
+  wasmFilename: string;
+  allowedHosts: string[];
+  maxHttpResponseBytes: number;
+  transformResponse: (response: any) => any;
+};
+
+export const createPluginInstance = async (
+  params: Omit<CreateToolClassParams, "transformResponse">,
+): Promise<Plugin> => {
+  const { wasmFilename, allowedHosts, maxHttpResponseBytes } = params;
+  const plugin = await createPlugin(`${WASM_DIRECTORY}/${wasmFilename}`, {
+    useWasi: true,
+    runInWorker: true,
+    allowedHosts,
+    memory: { maxHttpResponseBytes },
+  });
+  return plugin;
+};
+
+export const DEFAULT_TOOL_PARAMS: Omit<CreateToolClassParams, "wasmFilename"> =
+  {
+    allowedHosts: ["*"],
+    maxHttpResponseBytes: DEFAULT_MAX_HTTP_RESPONSE_BYTES,
+    transformResponse: (response: any) => response,
+  };
+
+export class ExtismToolFactory {
+  static async createToolClass(
+    toolName: string,
+    params: Omit<CreateToolClassParams, "wasmFilename"> = DEFAULT_TOOL_PARAMS,
+  ): Promise<new (params?: ToolClassParams) => BaseToolWithCall<ToolParams>> {
+    const config = { ...params, wasmFilename: `${toolName}.wasm` };
+    const plugin = await createPluginInstance(config);
+    try {
+      const wasmMetadata = await plugin.call("getMetadata");
+      if (!wasmMetadata) {
+        throw new Error("The WASM plugin did not return metadata.");
+      }
+      const defaultMetadata = wasmMetadata.json();
+
+      return class implements BaseToolWithCall<ToolParams> {
+        metadata: ToolMetadata<JSONSchemaType<ToolParams>>;
+
+        constructor(params?: ToolClassParams) {
+          this.metadata = params?.metadata || defaultMetadata;
+        }
+
+        async call(input: ToolParams): Promise<string> {
+          const pluginInstance = await createPluginInstance(config);
+          const data = await pluginInstance.call("call", JSON.stringify(input));
+          if (!data) return "No result";
+          const result = config.transformResponse(data.json());
+          await pluginInstance.close();
+          return result;
+        }
+      };
+    } catch (e) {
+      console.error(e);
+      throw new Error("Failed to create Tool instance.");
+    } finally {
+      await plugin.close();
+    }
+  }
+}

packages/extism-tools/src/index.ts

@@ -0,0 +1 @@
+export { ExtismToolFactory } from "./ExtismToolFactory.js";

packages/extism-tools/tsconfig.json

@@ -0,0 +1,21 @@
+{
+  "extends": "../../tsconfig.json",
+  "compilerOptions": {
+    "rootDir": "./src",
+    "outDir": "./dist",
+    "tsBuildInfoFile": "./dist/.tsbuildinfo",
+    "emitDeclarationOnly": true,
+    "module": "node16",
+    "moduleResolution": "node16",
+    "skipLibCheck": true,
+    "strict": true,
+    "types": ["node"]
+  },
+  "include": ["./src"],
+  "exclude": ["node_modules"],
+  "references": [
+    {
+      "path": "../env/tsconfig.json"
+    }
+  ]
+}

packages/extism-tools/wasm/index.d.ts

@@ -0,0 +1,4 @@
+declare module "main" {
+  export function getMetadata(): I32;
+  export function call(): I32;
+}

packages/extism-tools/wasm/todo.js

@@ -0,0 +1,34 @@
+function getMetadata() {
+  const metadata = {
+    name: "todo_tool",
+    description: "A tool helps search todo.",
+    parameters: {
+      type: "object",
+      properties: {
+        index: {
+          type: "number",
+          description: "The index of the todo to search for.",
+        },
+      },
+      required: ["index"],
+    },
+  };
+  Host.outputString(JSON.stringify(metadata));
+}
+
+function call() {
+  const params = JSON.parse(Host.inputString());
+  const index = params?.index;
+  if (!index) throw new Error("No index provided");
+  const request = {
+    method: "GET",
+    url: `https://jsonplaceholder.typicode.com/todos/${encodeURIComponent(index)}`,
+  };
+  const response = Http.request(request);
+  if (response.status != 200)
+    throw new Error(`Got non 200 response ${response.status}`);
+
+  Host.outputString(response.body);
+}
+
+module.exports = { getMetadata, call };

packages/extism-tools/wasm/wiki.js

@@ -0,0 +1,33 @@
+function getMetadata() {
+  const metadata = {
+    name: "wikipedia_tool",
+    description: "A tool that uses a query engine to search Wikipedia.",
+    parameters: {
+      type: "object",
+      properties: {
+        query: {
+          type: "string",
+          description: "The query to search for",
+        },
+      },
+      required: ["query"],
+    },
+  };
+  Host.outputString(JSON.stringify(metadata));
+}
+
+function call() {
+  const params = JSON.parse(Host.inputString());
+  const query = params?.query;
+  if (!query) throw new Error("No query provided");
+  const request = {
+    method: "GET",
+    url: `https://en.wikipedia.org/api/rest_v1/page/summary/${encodeURIComponent(query)}`,
+  };
+  const response = Http.request(request);
+  if (response.status != 200)
+    throw new Error(`Got non 200 response ${response.status}`);
+  Host.outputString(response.body);
+}
+
+module.exports = { getMetadata, call };

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

@@ -16,7 +16,7 @@
     "@cloudflare/workers-types": "^4.20240821.1",
     "@vitest/runner": "1.5.3",
     "@vitest/snapshot": "1.5.3",
-    "typescript": "^5.5.4",
+    "typescript": "^5.6.2",
     "vitest": "1.5.3",
     "wrangler": "^3.73.0"
   },

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

@@ -23,6 +23,6 @@
     "eslint-config-next": "14.2.7",
     "postcss": "^8.4.41",
     "tailwindcss": "^3.4.10",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   }
 }

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

@@ -17,6 +17,6 @@
     "@types/node": "^22.5.1",
     "@types/react": "^18.3.5",
     "@types/react-dom": "^18.3.0",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   }
 }

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

@@ -22,6 +22,6 @@
     "eslint-config-next": "14.2.7",
     "postcss": "^8.4.41",
     "tailwindcss": "^3.4.10",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   }
 }

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

@@ -20,7 +20,7 @@
     "@types/react-dom": "18.3.0",
     "autoprefixer": "10.4.20",
     "tailwindcss": "3.4.10",
-    "typescript": "5.5.4",
+    "typescript": "5.6.2",
     "vite-plugin-wasm": "^3.3.0"
   }
 }

packages/llamaindex/package.json

@@ -94,7 +94,7 @@
     "glob": "^11.0.0",
     "pg": "^8.12.0",
     "pgvector": "0.2.0",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   },
   "engines": {
     "node": ">=18.0.0"

packages/llamaindex/src/embeddings/JinaAIEmbedding.ts

@@ -8,10 +8,20 @@ function isLocal(url: ImageType): boolean {
   return new URL(url).protocol === "file:";
 }
 
+type TaskType =
+  | "retrieval.passage"
+  | "retrieval.query"
+  | "separation"
+  | "classification"
+  | "text-matching";
+type EncodingType = "float" | "binary" | "ubinary";
+
 export type JinaEmbeddingRequest = {
   input: Array<{ text: string } | { url: string } | { bytes: string }>;
   model?: string;
-  encoding_type?: "float" | "binary" | "ubinary";
+  encoding_type?: EncodingType;
+  task_type?: TaskType;
+  dimensions?: number;
 };
 
 export type JinaEmbeddingResponse = {
@@ -34,6 +44,9 @@ export class JinaAIEmbedding extends MultiModalEmbedding {
   apiKey: string;
   model: string;
   baseURL: string;
+  taskType: TaskType | undefined;
+  encodingType?: EncodingType | undefined;
+  dimensions?: number | undefined;
 
   async getTextEmbedding(text: string): Promise<number[]> {
     const result = await this.getJinaEmbedding({ input: [{ text }] });
@@ -71,9 +84,11 @@ export class JinaAIEmbedding extends MultiModalEmbedding {
       );
     }
     this.apiKey = apiKey;
-    this.model = init?.model ?? "jina-embeddings-v2-base-en";
+    this.model = init?.model ?? "jina-embeddings-v3";
     this.baseURL = init?.baseURL ?? "https://api.jina.ai/v1/embeddings";
     init?.embedBatchSize && (this.embedBatchSize = init?.embedBatchSize);
+    this.taskType = init?.taskType;
+    this.encodingType = init?.encodingType;
   }
 
   private async getImageInput(
@@ -89,11 +104,11 @@ export class JinaAIEmbedding extends MultiModalEmbedding {
   }
 
   private async getJinaEmbedding(
-    input: JinaEmbeddingRequest,
+    params: JinaEmbeddingRequest,
   ): Promise<JinaEmbeddingResponse> {
     // if input includes image, check if model supports multimodal embeddings
     if (
-      input.input.some((i) => "url" in i || "bytes" in i) &&
+      params.input.some((i) => "url" in i || "bytes" in i) &&
       !JINA_MULTIMODAL_MODELS.includes(this.model)
     ) {
       throw new Error(
@@ -109,13 +124,17 @@ export class JinaAIEmbedding extends MultiModalEmbedding {
       },
       body: JSON.stringify({
         model: this.model,
-        encoding_type: "float",
-        ...input,
+        encoding_type: this.encodingType ?? "float",
+        ...(this.taskType && { task_type: this.taskType }),
+        ...(this.dimensions !== undefined && { dimensions: this.dimensions }),
+        ...params,
       }),
     });
     if (!response.ok) {
+      const reason = await response.text();
+
       throw new Error(
-        `Request ${this.baseURL} failed with status ${response.status}`,
+        `Request failed with status ${response.status}: ${reason}`,
       );
     }
     const result: JinaEmbeddingResponse = await response.json();

packages/llamaindex/src/llm/portkey.ts

@@ -71,8 +71,8 @@ export class Portkey extends BaseLLM {
     this.baseURL = baseURL;
     this.session = getPortkeySession({
       ...rest,
-      apiKey: this.apiKey,
-      baseURL: this.baseURL,
+      apiKey: this.apiKey ?? null,
+      baseURL: this.baseURL ?? null,
     });
   }
 

packages/llamaindex/src/readers/LlamaParseReader.ts

@@ -245,13 +245,13 @@ export class LlamaParseReader extends FileReader {
   ) {
     super();
     Object.assign(this, params);
-    params.apiKey = params.apiKey ?? getEnv("LLAMA_CLOUD_API_KEY");
-    if (!params.apiKey) {
+    const apiKey = params.apiKey ?? getEnv("LLAMA_CLOUD_API_KEY");
+    if (!apiKey) {
       throw new Error(
         "API Key is required for LlamaParseReader. Please pass the apiKey parameter or set the LLAMA_CLOUD_API_KEY environment variable.",
       );
     }
-    this.apiKey = params.apiKey;
+    this.apiKey = apiKey;
 
     if (params.gpt4oMode) {
       params.gpt4oApiKey =

packages/llamaindex/tests/import.test.ts

@@ -1,11 +0,0 @@
-import { expect, test } from "vitest";
-
-test("Node classes should be included in the top level", async () => {
-  const { Document, IndexNode, TextNode, BaseNode } = await import(
-    "llamaindex"
-  );
-  expect(Document).toBeDefined();
-  expect(IndexNode).toBeDefined();
-  expect(TextNode).toBeDefined();
-  expect(BaseNode).toBeDefined();
-});

packages/wasm-tools/package.json

@@ -11,7 +11,7 @@
     "@swc/cli": "^0.4.0",
     "@swc/core": "^1.7.22",
     "assemblyscript": "^0.27.27",
-    "typescript": "^5.5.4"
+    "typescript": "^5.6.2"
   },
   "engines": {
     "node": ">=18.0.0"

pnpm-workspace.yaml

@@ -9,3 +9,4 @@ packages:
   - "packages/autotool/examples/*"
   - "examples/"
   - "examples/*"
+  - "packages/extism-tools/examples"

tsconfig.json

@@ -82,6 +82,12 @@
     },
     {
       "path": "./packages/experimental/tsconfig.json"
+    },
+    {
+      "path": "./packages/extism-tools/tsconfig.json"
+    },
+    {
+      "path": "./packages/extism-tools/examples/tsconfig.json"
     }
   ]
 }