RELEASING: Releasing 1 package(s)

Releases:
  create-llama@0.0.19

[skip ci]

.changeset/curly-mangos-brake.md

@@ -0,0 +1,5 @@
+---
+"llamaindex": patch
+---
+
+update dependencies

.husky/pre-commit

@@ -1,6 +1,3 @@
-#!/usr/bin/env sh
-. "$(dirname -- "$0")/_/husky.sh"
-
 pnpm format
 pnpm lint
 npx lint-staged

.husky/pre-push

@@ -1,4 +1 @@
-#!/usr/bin/env sh
-. "$(dirname -- "$0")/_/husky.sh"
-
 pnpm test

.npmrc

@@ -1 +1,2 @@
 auto-install-peers = true
+enable-pre-post-scripts = true

apps/docs/CHANGELOG.md

@@ -0,0 +1,7 @@
+# docs
+
+## 0.0.1
+
+### Patch Changes
+
+- 3154f52: chore: add qdrant readme

apps/docs/docs/modules/documents_and_nodes.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 0
+sidebar_position: 1
 ---
 
 # Documents and Nodes

apps/docs/docs/modules/llm.md

@@ -1,5 +1,5 @@
 ---
-sidebar_position: 0
+sidebar_position: 1
 ---
 
 # LLM

apps/docs/docs/modules/vectorStores/_category_.yml

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

apps/docs/docs/modules/vectorStores/qdrant.md

@@ -0,0 +1,88 @@
+# Qdrant Vector Store
+
+To run this example, you need to have a Qdrant instance running. You can run it with Docker:
+
+```bash
+docker pull qdrant/qdrant
+docker run -p 6333:6333 qdrant/qdrant
+```
+
+## Importing the modules
+
+```ts
+import fs from "node:fs/promises";
+import { Document, VectorStoreIndex, QdrantVectorStore } from "llamaindex";
+```
+
+## Load the documents
+
+```ts
+const path = "node_modules/llamaindex/examples/abramov.txt";
+const essay = await fs.readFile(path, "utf-8");
+```
+
+## Setup Qdrant
+
+```ts
+const vectorStore = new QdrantVectorStore({
+  url: "http://localhost:6333",
+  port: 6333,
+});
+```
+
+## Setup the index
+
+```ts
+const document = new Document({ text: essay, id_: path });
+
+const index = await VectorStoreIndex.fromDocuments([document], {
+  vectorStore,
+});
+```
+
+## Query the index
+
+```ts
+const queryEngine = index.asQueryEngine();
+
+const response = await queryEngine.query({
+  query: "What did the author do in college?",
+});
+
+// Output response
+console.log(response.toString());
+```
+
+## Full code
+
+```ts
+import fs from "node:fs/promises";
+import { Document, VectorStoreIndex, QdrantVectorStore } from "llamaindex";
+
+async function main() {
+  const path = "node_modules/llamaindex/examples/abramov.txt";
+  const essay = await fs.readFile(path, "utf-8");
+
+  const vectorStore = new QdrantVectorStore({
+    url: "http://localhost:6333",
+    port: 6333,
+  });
+
+  const document = new Document({ text: essay, id_: path });
+
+  const index = await VectorStoreIndex.fromDocuments([document], {
+    vectorStore,
+  });
+
+  const queryEngine = index.asQueryEngine();
+
+  const response = await queryEngine.query({
+    query: "What did the author do in college?",
+  });
+
+  // Output response
+  console.log(response.toString());
+}
+
+main().catch(console.error);
+```

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docs",
-  "version": "0.0.0",
+  "version": "0.0.1",
   "private": true,
   "scripts": {
     "docusaurus": "docusaurus",

examples/vectorIndexEmbed3.ts

@@ -0,0 +1,41 @@
+import fs from "node:fs/promises";
+
+import {
+  Document,
+  OpenAIEmbedding,
+  VectorStoreIndex,
+  serviceContextFromDefaults,
+} from "llamaindex";
+
+async function main() {
+  // Load essay from abramov.txt in Node
+  const path = "node_modules/llamaindex/examples/abramov.txt";
+
+  const essay = await fs.readFile(path, "utf-8");
+
+  // Create Document object with essay
+  const document = new Document({ text: essay, id_: path });
+
+  // Create service context and specify text-embedding-3-large
+  const embedModel = new OpenAIEmbedding({
+    model: "text-embedding-3-large",
+    dimensions: 1024,
+  });
+  const serviceContext = serviceContextFromDefaults({ embedModel });
+
+  // Split text and create embeddings. Store them in a VectorStoreIndex
+  const index = await VectorStoreIndex.fromDocuments([document], {
+    serviceContext,
+  });
+
+  // Query the index
+  const queryEngine = index.asQueryEngine();
+  const response = await queryEngine.query({
+    query: "What did the author do in college?",
+  });
+
+  // Output response
+  console.log(response.toString());
+}
+
+main().catch(console.error);

package.json

@@ -8,7 +8,7 @@
     "format": "prettier --ignore-unknown --cache --check .",
     "format:write": "prettier --ignore-unknown --write .",
     "lint": "turbo run lint",
-    "prepare": "husky install",
+    "prepare": "husky",
     "test": "turbo run test",
     "type-check": "tsc -b --diagnostics",
     "release": "pnpm run build:release && changeset publish",
@@ -18,20 +18,20 @@
   },
   "devDependencies": {
     "@changesets/cli": "^2.27.1",
-    "@turbo/gen": "^1.11.2",
+    "@turbo/gen": "^1.11.3",
     "@types/jest": "^29.5.11",
     "eslint": "^8.56.0",
     "eslint-config-custom": "workspace:*",
-    "husky": "^8.0.3",
+    "husky": "^9.0.6",
     "jest": "^29.7.0",
     "lint-staged": "^15.2.0",
     "prettier": "^3.2.4",
     "prettier-plugin-organize-imports": "^3.2.4",
-    "ts-jest": "^29.1.1",
-    "turbo": "^1.11.2",
+    "ts-jest": "^29.1.2",
+    "turbo": "^1.11.3",
     "typescript": "^5.3.3"
   },
-  "packageManager": "pnpm@8.10.5+sha256.a4bd9bb7b48214bbfcd95f264bd75bb70d100e5d4b58808f5cd6ab40c6ac21c5",
+  "packageManager": "pnpm@8.14.3+sha256.2d0363bb6c314daa67087ef07743eea1ba2e2d360c835e8fec6b5575e4ed9484",
   "pnpm": {
     "overrides": {
       "trim": "1.0.1",

packages/core/CHANGELOG.md

@@ -1,5 +1,29 @@
 # llamaindex
 
+## 0.1.2
+
+- e4b807a: fix: invalid package.json
+
+## 0.1.1
+
+No changes for this release.
+
+## 0.1.0
+
+### Minor Changes
+
+- 3154f52: chore: add qdrant readme
+
+### Patch Changes
+
+- bb66cb7: add new OpenAI embeddings (with dimension reduction support)
+
+## 0.0.51
+
+### Patch Changes
+
+- fda8024: revert: export conditions not working with moduleResolution `node`
+
 ## 0.0.50
 
 ### Patch Changes

packages/core/README.md

@@ -1,130 +0,0 @@
-# LlamaIndex.TS
-
-[![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 Typescript and Javascript.
-
-Documentation: https://ts.llamaindex.ai/
-
-## What is LlamaIndex.TS?
-
-LlamaIndex.TS aims to be a lightweight, easy to use set of libraries to help you integrate large language models into your applications with your own data.
-
-## Getting started with an example:
-
-LlamaIndex.TS requires Node v18 or higher. You can download it from https://nodejs.org or use https://nvm.sh (our preferred option).
-
-In a new folder:
-
-```bash
-export OPENAI_API_KEY="sk-......" # Replace with your key from https://platform.openai.com/account/api-keys
-pnpm init
-pnpm install typescript
-pnpm exec tsc --init # if needed
-pnpm install llamaindex
-pnpm install @types/node
-```
-
-Create the file example.ts
-
-```ts
-// example.ts
-import fs from "fs/promises";
-import { Document, VectorStoreIndex } from "llamaindex";
-
-async function main() {
-  // Load essay from abramov.txt in Node
-  const essay = await fs.readFile(
-    "node_modules/llamaindex/examples/abramov.txt",
-    "utf-8",
-  );
-
-  // Create Document object with essay
-  const document = new Document({ text: essay });
-
-  // Split text and create embeddings. Store them in a VectorStoreIndex
-  const index = await VectorStoreIndex.fromDocuments([document]);
-
-  // Query the index
-  const queryEngine = index.asQueryEngine();
-  const response = await queryEngine.query(
-    "What did the author do in college?",
-  );
-
-  // Output response
-  console.log(response.toString());
-}
-
-main();
-```
-
-Then you can run it using
-
-```bash
-pnpx ts-node example.ts
-```
-
-## Playground
-
-Check out our NextJS playground at https://llama-playground.vercel.app/. The source is available at https://github.com/run-llama/ts-playground
-
-## Core concepts for getting started:
-
-- [Document](/packages/core/src/Node.ts): A document represents a text file, PDF file or other contiguous piece of data.
-
-- [Node](/packages/core/src/Node.ts): The basic data building block. Most commonly, these are parts of the document split into manageable pieces that are small enough to be fed into an embedding model and LLM.
-
-- [Embedding](/packages/core/src/Embedding.ts): Embeddings are sets of floating point numbers which represent the data in a Node. By comparing the similarity of embeddings, we can derive an understanding of the similarity of two pieces of data. One use case is to compare the embedding of a question with the embeddings of our Nodes to see which Nodes may contain the data needed to answer that quesiton.
-
-- [Indices](/packages/core/src/indices/): Indices store the Nodes and the embeddings of those nodes. QueryEngines retrieve Nodes from these Indices using embedding similarity.
-
-- [QueryEngine](/packages/core/src/QueryEngine.ts): Query engines are what generate the query you put in and give you back the result. Query engines generally combine a pre-built prompt with selected Nodes from your Index to give the LLM the context it needs to answer your query.
-
-- [ChatEngine](/packages/core/src/ChatEngine.ts): A ChatEngine helps you build a chatbot that will interact with your Indices.
-
-- [SimplePrompt](/packages/core/src/Prompt.ts): A simple standardized function call definition that takes in inputs and formats them in a template literal. SimplePrompts can be specialized using currying and combined using other SimplePrompt functions.
-
-## Note: NextJS:
-
-If you're using NextJS App Router, you'll need to use the NodeJS runtime (default) and add the following config to your next.config.js to have it use imports/exports in the same way Node does.
-
-```js
-export const runtime = "nodejs"; // default
-```
-
-```js
-// next.config.js
-/** @type {import('next').NextConfig} */
-const nextConfig = {
-  webpack: (config) => {
-    config.resolve.alias = {
-      ...config.resolve.alias,
-      sharp$: false,
-      "onnxruntime-node$": false,
-    };
-    return config;
-  },
-};
-
-module.exports = nextConfig;
-```
-
-## Supported LLMs:
-
-- OpenAI GPT-3.5-turbo and GPT-4
-- Anthropic Claude Instant and Claude 2
-- Llama2 Chat LLMs (70B, 13B, and 7B parameters)
-- MistralAI Chat LLMs
-
-## Contributing:
-
-We are in the very early days of LlamaIndex.TS. If you’re interested in hacking on it with us check out our [contributing guide](/CONTRIBUTING.md)
-
-## Bugs? Questions?
-
-Please join our Discord! https://discord.com/invite/eN6D2HQ4aX

packages/core/package.json

@@ -1,16 +1,17 @@
 {
   "name": "llamaindex",
-  "version": "0.0.50",
+  "private": true,
+  "version": "0.1.2",
   "license": "MIT",
   "dependencies": {
-    "@anthropic-ai/sdk": "^0.9.1",
-    "@datastax/astra-db-ts": "^0.1.2",
-    "@mistralai/mistralai": "^0.0.7",
+    "@anthropic-ai/sdk": "^0.12.4",
+    "@datastax/astra-db-ts": "^0.1.4",
+    "@mistralai/mistralai": "^0.0.10",
     "@notionhq/client": "^2.2.14",
-    "@pinecone-database/pinecone": "^1.1.2",
+    "@pinecone-database/pinecone": "^1.1.3",
     "@qdrant/js-client-rest": "^1.7.0",
-    "@xenova/transformers": "^2.10.0",
-    "assemblyai": "^4.0.0",
+    "@xenova/transformers": "^2.14.1",
+    "assemblyai": "^4.2.1",
     "chromadb": "~1.7.3",
     "file-type": "^18.7.0",
     "js-tiktoken": "^1.0.8",
@@ -19,26 +20,28 @@
     "md-utils-ts": "^2.0.0",
     "mongodb": "^6.3.0",
     "notion-md-crawler": "^0.0.2",
-    "openai": "^4.20.1",
+    "openai": "^4.26.0",
     "papaparse": "^5.4.1",
     "pathe": "^1.1.2",
     "pdfjs-dist": "4.0.269",
     "pg": "^8.11.3",
-    "pgvector": "^0.1.5",
+    "pgvector": "^0.1.7",
     "portkey-ai": "^0.1.16",
     "rake-modified": "^1.0.8",
-    "replicate": "^0.21.1",
-    "string-strip-html": "^13.4.3",
+    "replicate": "^0.25.2",
+    "string-strip-html": "^13.4.5",
     "wink-nlp": "^1.14.3"
   },
   "devDependencies": {
     "@aws-crypto/sha256-js": "^5.2.0",
+    "@types/edit-json-file": "^1.7.3",
     "@types/jest": "^29.5.11",
     "@types/lodash": "^4.14.202",
-    "@types/node": "^18.19.6",
+    "@types/node": "^18.19.9",
     "@types/papaparse": "^5.3.14",
-    "@types/pg": "^8.10.9",
-    "bunchee": "^4.4.1",
+    "@types/pg": "^8.11.0",
+    "bunchee": "^4.4.2",
+    "edit-json-file": "^1.8.0",
     "madge": "^6.1.0",
     "typescript": "^5.3.3"
   },
@@ -50,154 +53,19 @@
   "exports": {
     ".": {
       "types": "./dist/index.d.mts",
-      "edge-light": "./dist/index.edge-light.mjs",
       "import": "./dist/index.mjs",
+      "edge-light": "./dist/index.mjs",
       "require": "./dist/index.js"
     },
     "./env": {
       "types": "./dist/env.d.mts",
-      "edge-light": "./dist/env.edge-light.mjs",
       "import": "./dist/env.mjs",
+      "edge-light": "./dist/env.mjs",
       "require": "./dist/env.js"
-    },
-    "./storage/FileSystem": {
-      "types": "./dist/storage/FileSystem.d.mts",
-      "edge-light": "./dist/storage/FileSystem.edge-light.mjs",
-      "import": "./dist/storage/FileSystem.mjs",
-      "require": "./dist/storage/FileSystem.js"
-    },
-    "./ChatHistory": {
-      "types": "./dist/ChatHistory.d.mts",
-      "import": "./dist/ChatHistory.mjs",
-      "require": "./dist/ChatHistory.js"
-    },
-    "./constants": {
-      "types": "./dist/constants.d.mts",
-      "import": "./dist/constants.mjs",
-      "require": "./dist/constants.js"
-    },
-    "./GlobalsHelper": {
-      "types": "./dist/GlobalsHelper.d.mts",
-      "import": "./dist/GlobalsHelper.mjs",
-      "require": "./dist/GlobalsHelper.js"
-    },
-    "./Node": {
-      "types": "./dist/Node.d.mts",
-      "import": "./dist/Node.mjs",
-      "require": "./dist/Node.js"
-    },
-    "./OutputParser": {
-      "types": "./dist/OutputParser.d.mts",
-      "import": "./dist/OutputParser.mjs",
-      "require": "./dist/OutputParser.js"
-    },
-    "./Prompt": {
-      "types": "./dist/Prompt.d.mts",
-      "import": "./dist/Prompt.mjs",
-      "require": "./dist/Prompt.js"
-    },
-    "./PromptHelper": {
-      "types": "./dist/PromptHelper.d.mts",
-      "import": "./dist/PromptHelper.mjs",
-      "require": "./dist/PromptHelper.js"
-    },
-    "./QueryEngine": {
-      "types": "./dist/QueryEngine.d.mts",
-      "import": "./dist/QueryEngine.mjs",
-      "require": "./dist/QueryEngine.js"
-    },
-    "./QuestionGenerator": {
-      "types": "./dist/QuestionGenerator.d.mts",
-      "import": "./dist/QuestionGenerator.mjs",
-      "require": "./dist/QuestionGenerator.js"
-    },
-    "./Response": {
-      "types": "./dist/Response.d.mts",
-      "import": "./dist/Response.mjs",
-      "require": "./dist/Response.js"
-    },
-    "./Retriever": {
-      "types": "./dist/Retriever.d.mts",
-      "import": "./dist/Retriever.mjs",
-      "require": "./dist/Retriever.js"
-    },
-    "./ServiceContext": {
-      "types": "./dist/ServiceContext.d.mts",
-      "import": "./dist/ServiceContext.mjs",
-      "require": "./dist/ServiceContext.js"
-    },
-    "./TextSplitter": {
-      "types": "./dist/TextSplitter.d.mts",
-      "import": "./dist/TextSplitter.mjs",
-      "require": "./dist/TextSplitter.js"
-    },
-    "./Tool": {
-      "types": "./dist/Tool.d.mts",
-      "import": "./dist/Tool.mjs",
-      "require": "./dist/Tool.js"
-    },
-    "./readers/AssemblyAI": {
-      "types": "./dist/readers/AssemblyAI.d.mts",
-      "import": "./dist/readers/AssemblyAI.mjs",
-      "require": "./dist/readers/AssemblyAI.js"
-    },
-    "./readers/base": {
-      "types": "./dist/readers/base.d.mts",
-      "import": "./dist/readers/base.mjs",
-      "require": "./dist/readers/base.js"
-    },
-    "./readers/CSVReader": {
-      "types": "./dist/readers/CSVReader.d.mts",
-      "import": "./dist/readers/CSVReader.mjs",
-      "require": "./dist/readers/CSVReader.js"
-    },
-    "./readers/DocxReader": {
-      "types": "./dist/readers/DocxReader.d.mts",
-      "import": "./dist/readers/DocxReader.mjs",
-      "require": "./dist/readers/DocxReader.js"
-    },
-    "./readers/HTMLReader": {
-      "types": "./dist/readers/HTMLReader.d.mts",
-      "import": "./dist/readers/HTMLReader.mjs",
-      "require": "./dist/readers/HTMLReader.js"
-    },
-    "./readers/ImageReader": {
-      "types": "./dist/readers/ImageReader.d.mts",
-      "import": "./dist/readers/ImageReader.mjs",
-      "require": "./dist/readers/ImageReader.js"
-    },
-    "./readers/MarkdownReader": {
-      "types": "./dist/readers/MarkdownReader.d.mts",
-      "import": "./dist/readers/MarkdownReader.mjs",
-      "require": "./dist/readers/MarkdownReader.js"
-    },
-    "./readers/NotionReader": {
-      "types": "./dist/readers/NotionReader.d.mts",
-      "import": "./dist/readers/NotionReader.mjs",
-      "require": "./dist/readers/NotionReader.js"
-    },
-    "./readers/PDFReader": {
-      "types": "./dist/readers/PDFReader.d.mts",
-      "import": "./dist/readers/PDFReader.mjs",
-      "require": "./dist/readers/PDFReader.js"
-    },
-    "./readers/SimpleDirectoryReader": {
-      "types": "./dist/readers/SimpleDirectoryReader.d.mts",
-      "import": "./dist/readers/SimpleDirectoryReader.mjs",
-      "require": "./dist/readers/SimpleDirectoryReader.js"
-    },
-    "./readers/SimpleMongoReader": {
-      "types": "./dist/readers/SimpleMongoReader.d.mts",
-      "import": "./dist/readers/SimpleMongoReader.mjs",
-      "require": "./dist/readers/SimpleMongoReader.js"
     }
   },
   "files": [
-    "dist",
-    "examples",
-    "src",
-    "types",
-    "CHANGELOG.md"
+    "**"
   ],
   "repository": {
     "type": "git",
@@ -208,6 +76,10 @@
     "lint": "eslint .",
     "test": "jest",
     "build": "bunchee",
+    "postbuild": "pnpm run copy && pnpm run modify-package-json",
+    "copy": "cp -r package.json CHANGELOG.md ../../README.md ../../LICENSE examples src dist/",
+    "modify-package-json": "node ./scripts/modify-package-json.mjs",
+    "prepublish": "pnpm run modify-package-json && echo \"please cd ./dist and run pnpm publish\" && exit 1",
     "dev": "bunchee -w",
     "circular-check": "madge --circular ./src/*.ts"
   }

packages/core/scripts/modify-package-json.mjs

@@ -0,0 +1,25 @@
+#!/usr/bin/env node
+/**
+ * This script is used to modify the package.json file in the dist folder
+ * so that it can be published to npm.
+ */
+import editJsonFile from "edit-json-file";
+import fs from "node:fs/promises";
+
+{
+  await fs.copyFile("./package.json", "./dist/package.json");
+  const file = editJsonFile("./dist/package.json");
+
+  file.unset("scripts");
+  file.unset("private");
+  await new Promise((resolve) => file.save(resolve));
+}
+{
+  const packageJson = await fs.readFile("./dist/package.json", "utf8");
+  const modifiedPackageJson = packageJson.replaceAll("./dist/", "./");
+  await fs.writeFile(
+    "./dist/package.json",
+    JSON.stringify(JSON.parse(modifiedPackageJson), null, 2),
+    "utf8",
+  );
+}

packages/core/src/constants.ts

@@ -6,6 +6,4 @@ export const DEFAULT_CHUNK_OVERLAP = 20;
 export const DEFAULT_CHUNK_OVERLAP_RATIO = 0.1;
 export const DEFAULT_SIMILARITY_TOP_K = 2;
 
-// NOTE: for text-embedding-ada-002
-export const DEFAULT_EMBEDDING_DIM = 1536;
 export const DEFAULT_PADDING = 5;

packages/core/src/embeddings/OpenAIEmbedding.ts

@@ -9,28 +9,55 @@ import {
 import { OpenAISession, getOpenAISession } from "../llm/openai";
 import { BaseEmbedding } from "./types";
 
-export enum OpenAIEmbeddingModelType {
-  TEXT_EMBED_ADA_002 = "text-embedding-ada-002",
-}
+export const ALL_OPENAI_EMBEDDING_MODELS = {
+  "text-embedding-ada-002": {
+    dimensions: 1536,
+    maxTokens: 8191,
+  },
+  "text-embedding-3-small": {
+    dimensions: 1536,
+    dimensionOptions: [512, 1536],
+    maxTokens: 8191,
+  },
+  "text-embedding-3-large": {
+    dimensions: 3072,
+    dimensionOptions: [256, 1024, 3072],
+    maxTokens: 8191,
+  },
+};
 
 export class OpenAIEmbedding extends BaseEmbedding {
-  model: OpenAIEmbeddingModelType | string;
+  /** embeddding model. defaults to "text-embedding-ada-002" */
+  model: string;
+  /** number of dimensions of the resulting vector, for models that support choosing fewer dimensions. undefined will default to model default */
+  dimensions: number | undefined;
 
   // OpenAI session params
+
+  /** api key */
   apiKey?: string = undefined;
+  /** maximum number of retries, default 10 */
   maxRetries: number;
+  /** timeout in ms, default 60 seconds  */
   timeout?: number;
+  /** other session options for OpenAI */
   additionalSessionOptions?: Omit<
     Partial<OpenAIClientOptions>,
     "apiKey" | "maxRetries" | "timeout"
   >;
 
+  /** session object */
   session: OpenAISession;
 
+  /**
+   * OpenAI Embedding
+   * @param init - initial parameters
+   */
   constructor(init?: Partial<OpenAIEmbedding> & { azure?: AzureOpenAIConfig }) {
     super();
 
-    this.model = init?.model ?? OpenAIEmbeddingModelType.TEXT_EMBED_ADA_002;
+    this.model = init?.model ?? "text-embedding-ada-002";
+    this.dimensions = init?.dimensions; // if no dimensions provided, will be undefined/not sent to OpenAI
 
     this.maxRetries = init?.maxRetries ?? 10;
     this.timeout = init?.timeout ?? 60 * 1000; // Default is 60 seconds
@@ -76,6 +103,7 @@ export class OpenAIEmbedding extends BaseEmbedding {
   private async getOpenAIEmbedding(input: string) {
     const { data } = await this.session.openai.embeddings.create({
       model: this.model,
+      dimensions: this.dimensions, // only sent to OpenAI if set by user
       input,
     });
 

packages/core/src/llm/LLM.ts

@@ -41,14 +41,21 @@ import {
 export const GPT4_MODELS = {
   "gpt-4": { contextWindow: 8192 },
   "gpt-4-32k": { contextWindow: 32768 },
+  "gpt-4-32k-0613": { contextWindow: 32768 },
+  "gpt-4-turbo-preview": { contextWindow: 128000 },
   "gpt-4-1106-preview": { contextWindow: 128000 },
-  "gpt-4-vision-preview": { contextWindow: 8192 },
+  "gpt-4-0125-preview": { contextWindow: 128000 },
+  "gpt-4-vision-preview": { contextWindow: 128000 },
 };
 
+// NOTE we don't currently support gpt-3.5-turbo-instruct and don't plan to in the near future
 export const GPT35_MODELS = {
   "gpt-3.5-turbo": { contextWindow: 4096 },
+  "gpt-3.5-turbo-0613": { contextWindow: 4096 },
   "gpt-3.5-turbo-16k": { contextWindow: 16384 },
+  "gpt-3.5-turbo-16k-0613": { contextWindow: 16384 },
   "gpt-3.5-turbo-1106": { contextWindow: 16384 },
+  "gpt-3.5-turbo-0125": { contextWindow: 16384 },
 };
 
 /**

packages/core/src/llm/azure.ts

@@ -17,6 +17,14 @@ const ALL_AZURE_OPENAI_CHAT_MODELS = {
   },
   "gpt-4": { contextWindow: 8192, openAIModel: "gpt-4" },
   "gpt-4-32k": { contextWindow: 32768, openAIModel: "gpt-4-32k" },
+  "gpt-4-vision-preview": {
+    contextWindow: 128000,
+    openAIModel: "gpt-4-vision-preview",
+  },
+  "gpt-4-1106-preview": {
+    contextWindow: 128000,
+    openAIModel: "gpt-4-1106-preview",
+  },
 };
 
 const ALL_AZURE_OPENAI_EMBEDDING_MODELS = {
@@ -25,13 +33,29 @@ const ALL_AZURE_OPENAI_EMBEDDING_MODELS = {
     openAIModel: "text-embedding-ada-002",
     maxTokens: 8191,
   },
+  "text-embedding-3-small": {
+    dimensions: 1536,
+    dimensionOptions: [512, 1536],
+    openAIModel: "text-embedding-3-small",
+    maxTokens: 8191,
+  },
+  "text-embedding-3-large": {
+    dimensions: 3072,
+    dimensionOptions: [256, 1024, 3072],
+    openAIModel: "text-embedding-3-large",
+    maxTokens: 8191,
+  },
 };
 
 const ALL_AZURE_API_VERSIONS = [
   "2022-12-01",
   "2023-05-15",
-  "2023-06-01-preview",
-  "2023-07-01-preview",
+  "2023-03-15-preview", // retiring 2024-04-02
+  "2023-06-01-preview", // retiring 2024-04-02
+  "2023-07-01-preview", // retiring 2024-04-02
+  "2023-08-01-preview", // retiring 2024-04-02
+  "2023-09-01-preview",
+  "2023-12-01-preview",
 ];
 
 const DEFAULT_API_VERSION = "2023-05-15";

packages/core/src/storage/vectorStore/PGVectorStore.ts

@@ -20,6 +20,7 @@ export class PGVectorStore implements VectorStore {
   private schemaName: string = PGVECTOR_SCHEMA;
   private tableName: string = PGVECTOR_TABLE;
   private connectionString: string | undefined = undefined;
+  private dimensions: number = 1536;
 
   private db?: pg.Client;
 
@@ -38,15 +39,18 @@ export class PGVectorStore implements VectorStore {
    * @param {string} config.schemaName - The name of the schema (optional). Defaults to PGVECTOR_SCHEMA.
    * @param {string} config.tableName - The name of the table (optional). Defaults to PGVECTOR_TABLE.
    * @param {string} config.connectionString - The connection string (optional).
+   * @param {number} config.dimensions - The dimensions of the embedding model.
    */
   constructor(config?: {
     schemaName?: string;
     tableName?: string;
     connectionString?: string;
+    dimensions?: number;
   }) {
     this.schemaName = config?.schemaName ?? PGVECTOR_SCHEMA;
     this.tableName = config?.tableName ?? PGVECTOR_TABLE;
     this.connectionString = config?.connectionString;
+    this.dimensions = config?.dimensions ?? 1536;
   }
 
   /**
@@ -108,7 +112,7 @@ export class PGVectorStore implements VectorStore {
       collection VARCHAR,
       document TEXT,
       metadata JSONB DEFAULT '{}',
-      embeddings VECTOR(1536)
+      embeddings VECTOR(${this.dimensions})
     )`;
     await db.query(tbl);
 

packages/core/src/storage/vectorStore/QdrantVectorStore.ts

@@ -54,11 +54,9 @@ export class QdrantVectorStore implements VectorStore {
     apiKey,
     batchSize,
   }: QdrantParams) {
-    if (!client && (!url || !apiKey)) {
-      if (!url || !apiKey || !collectionName) {
-        throw new Error(
-          "QdrantVectorStore requires url, apiKey and collectionName",
-        );
+    if (!client && !url) {
+      if (!url || !collectionName) {
+        throw new Error("QdrantVectorStore requires url and collectionName");
       }
     }
 

packages/core/src/types/mistral.d.ts

@@ -1,3 +0,0 @@
-declare module "@mistralai/mistralai" {
-  export = MistralClient;
-}

packages/create-llama/CHANGELOG.md

@@ -1,5 +1,12 @@
 # create-llama
 
+## 0.0.19
+
+### Patch Changes
+
+- 3a29a80: Add node_modules to gitignore in Express backends
+- fe03aaa: feat: generate llama pack example
+
 ## 0.0.18
 
 ### Patch Changes

packages/create-llama/create-app.ts

@@ -32,6 +32,7 @@ export async function createApp({
   openAiKey,
   model,
   communityProjectPath,
+  llamapack,
   vectorDb,
   externalPort,
   postInstallAction,
@@ -75,6 +76,7 @@ export async function createApp({
     openAiKey,
     model,
     communityProjectPath,
+    llamapack,
     vectorDb,
     externalPort,
     postInstallAction,

packages/create-llama/helpers/constant.ts

@@ -1,2 +1,6 @@
 export const COMMUNITY_OWNER = "run-llama";
 export const COMMUNITY_REPO = "create_llama_projects";
+export const LLAMA_PACK_OWNER = "run-llama";
+export const LLAMA_PACK_REPO = "llama-hub";
+export const LLAMA_HUB_FOLDER_PATH = `${LLAMA_PACK_OWNER}/${LLAMA_PACK_REPO}/main/llama_hub`;
+export const LLAMA_PACK_CONFIG_PATH = `${LLAMA_HUB_FOLDER_PATH}/llama_packs/library.json`;

packages/create-llama/helpers/index.ts

@@ -7,6 +7,7 @@ import { cyan } from "picocolors";
 
 import { COMMUNITY_OWNER, COMMUNITY_REPO } from "./constant";
 import { PackageManager } from "./get-pkg-manager";
+import { installLlamapackProject } from "./llama-pack";
 import { isHavingPoetryLockFile, tryPoetryRun } from "./poetry";
 import { installPythonTemplate } from "./python";
 import { downloadAndExtractRepo } from "./repo";
@@ -153,6 +154,11 @@ export const installTemplate = async (
     return;
   }
 
+  if (props.template === "llamapack" && props.llamapack) {
+    await installLlamapackProject(props);
+    return;
+  }
+
   if (props.framework === "fastapi") {
     await installPythonTemplate(props);
   } else {

packages/create-llama/helpers/llama-pack.ts

@@ -0,0 +1,91 @@
+import fs from "fs/promises";
+import path from "path";
+import { LLAMA_HUB_FOLDER_PATH, LLAMA_PACK_CONFIG_PATH } from "./constant";
+import { copy } from "./copy";
+import { installPythonDependencies } from "./python";
+import { getRepoRawContent } from "./repo";
+import { InstallTemplateArgs } from "./types";
+
+export async function getAvailableLlamapackOptions(): Promise<
+  {
+    name: string;
+    folderPath: string;
+    example: boolean | undefined;
+  }[]
+> {
+  const libraryJsonRaw = await getRepoRawContent(LLAMA_PACK_CONFIG_PATH);
+  const libraryJson = JSON.parse(libraryJsonRaw);
+  const llamapackKeys = Object.keys(libraryJson);
+  return llamapackKeys
+    .map((key) => ({
+      name: key,
+      folderPath: libraryJson[key].id,
+      example: libraryJson[key].example,
+    }))
+    .filter((item) => !!item.example);
+}
+
+const copyLlamapackEmptyProject = async ({
+  root,
+}: Pick<InstallTemplateArgs, "root">) => {
+  const templatePath = path.join(
+    __dirname,
+    "..",
+    "templates/components/sample-projects/llamapack",
+  );
+  await copy("**", root, {
+    parents: true,
+    cwd: templatePath,
+  });
+};
+
+const copyData = async ({
+  root,
+}: Pick<InstallTemplateArgs, "root" | "llamapack">) => {
+  const dataPath = path.join(__dirname, "..", "templates/components/data");
+  await copy("**", path.join(root, "data"), {
+    parents: true,
+    cwd: dataPath,
+  });
+};
+
+const installLlamapackExample = async ({
+  root,
+  llamapack,
+}: Pick<InstallTemplateArgs, "root" | "llamapack">) => {
+  const exampleFileName = "example.py";
+  const readmeFileName = "README.md";
+  const exampleFilePath = `${LLAMA_HUB_FOLDER_PATH}/${llamapack}/${exampleFileName}`;
+  const readmeFilePath = `${LLAMA_HUB_FOLDER_PATH}/${llamapack}/${readmeFileName}`;
+
+  // Download example.py from llamapack and save to root
+  const exampleContent = await getRepoRawContent(exampleFilePath);
+  await fs.writeFile(path.join(root, exampleFileName), exampleContent);
+
+  // Download README.md from llamapack and combine with README-template.md,
+  // save to root and then delete template file
+  const readmeContent = await getRepoRawContent(readmeFilePath);
+  const readmeTemplateContent = await fs.readFile(
+    path.join(root, "README-template.md"),
+    "utf-8",
+  );
+  await fs.writeFile(
+    path.join(root, readmeFileName),
+    `${readmeContent}\n${readmeTemplateContent}`,
+  );
+  await fs.unlink(path.join(root, "README-template.md"));
+};
+
+export const installLlamapackProject = async ({
+  root,
+  llamapack,
+  postInstallAction,
+}: Pick<InstallTemplateArgs, "root" | "llamapack" | "postInstallAction">) => {
+  console.log("\nInstalling Llamapack project:", llamapack!);
+  await copyLlamapackEmptyProject({ root });
+  await copyData({ root });
+  await installLlamapackExample({ root, llamapack });
+  if (postInstallAction !== "none") {
+    installPythonDependencies({ noRoot: true });
+  }
+};

packages/create-llama/helpers/poetry.ts

@@ -10,9 +10,11 @@ export function isPoetryAvailable(): boolean {
   return false;
 }
 
-export function tryPoetryInstall(): boolean {
+export function tryPoetryInstall(noRoot: boolean): boolean {
   try {
-    execSync("poetry install", { stdio: "inherit" });
+    execSync(`poetry install${noRoot ? " --no-root" : ""}`, {
+      stdio: "inherit",
+    });
     return true;
   } catch (_) {}
   return false;

packages/create-llama/helpers/python.ts

@@ -92,12 +92,14 @@ export const addDependencies = async (
   }
 };
 
-export const installPythonDependencies = (root: string) => {
+export const installPythonDependencies = (
+  { noRoot }: { noRoot: boolean } = { noRoot: false },
+) => {
   if (isPoetryAvailable()) {
     console.log(
       `Installing python dependencies using poetry. This may take a while...`,
     );
-    const installSuccessful = tryPoetryInstall();
+    const installSuccessful = tryPoetryInstall(noRoot);
     if (!installSuccessful) {
       console.error(
         red("Install failed. Please install dependencies manually."),
@@ -181,6 +183,6 @@ export const installPythonTemplate = async ({
   await addDependencies(root, addOnDependencies);
 
   if (postInstallAction !== "none") {
-    installPythonDependencies(root);
+    installPythonDependencies();
   }
 };

packages/create-llama/helpers/repo.ts

@@ -61,3 +61,11 @@ export async function getRepoRootFolders(
   const folders = data.filter((item) => item.type === "dir");
   return folders.map((item) => item.name);
 }
+
+export async function getRepoRawContent(repoFilePath: string) {
+  const url = `https://raw.githubusercontent.com/${repoFilePath}`;
+  const response = await got(url, {
+    responseType: "text",
+  });
+  return response.body;
+}

packages/create-llama/helpers/types.ts

@@ -1,6 +1,6 @@
 import { PackageManager } from "../helpers/get-pkg-manager";
 
-export type TemplateType = "simple" | "streaming" | "community";
+export type TemplateType = "simple" | "streaming" | "community" | "llamapack";
 export type TemplateFramework = "nextjs" | "express" | "fastapi";
 export type TemplateEngine = "simple" | "context";
 export type TemplateUI = "html" | "shadcn";
@@ -23,6 +23,7 @@ export interface InstallTemplateArgs {
   forBackend?: string;
   model: string;
   communityProjectPath?: string;
+  llamapack?: string;
   vectorDb?: TemplateVectorDB;
   externalPort?: number;
   postInstallAction?: TemplatePostInstallAction;

packages/create-llama/index.ts

@@ -237,6 +237,7 @@ async function run(): Promise<void> {
     openAiKey: program.openAiKey,
     model: program.model,
     communityProjectPath: program.communityProjectPath,
+    llamapack: program.llamapack,
     vectorDb: program.vectorDb,
     externalPort: program.externalPort,
     postInstallAction: program.postInstallAction,

packages/create-llama/package.json

@@ -1,6 +1,6 @@
 {
   "name": "create-llama",
-  "version": "0.0.18",
+  "version": "0.0.19",
   "keywords": [
     "rag",
     "llamaindex",

packages/create-llama/questions.ts

@@ -7,6 +7,7 @@ import prompts from "prompts";
 import { InstallAppArgs } from "./create-app";
 import { TemplateFramework } from "./helpers";
 import { COMMUNITY_OWNER, COMMUNITY_REPO } from "./helpers/constant";
+import { getAvailableLlamapackOptions } from "./helpers/llama-pack";
 import { getRepoRootFolders } from "./helpers/repo";
 
 export type QuestionArgs = Omit<InstallAppArgs, "appPath" | "packageManager">;
@@ -37,6 +38,7 @@ const defaults: QuestionArgs = {
   openAiKey: "",
   model: "gpt-3.5-turbo",
   communityProjectPath: "",
+  llamapack: "",
   postInstallAction: "dependencies",
 };
 
@@ -129,6 +131,48 @@ export const askQuestions = async (
     field: K,
   ): QuestionArgs[K] => preferences[field] ?? defaults[field];
 
+  // Ask for next action after installation
+  async function askPostInstallAction() {
+    if (program.postInstallAction === undefined) {
+      if (ciInfo.isCI) {
+        program.postInstallAction = getPrefOrDefault("postInstallAction");
+      } else {
+        let actionChoices = [
+          {
+            title: "Just generate code (~1 sec)",
+            value: "none",
+          },
+          {
+            title: "Generate code and install dependencies (~2 min)",
+            value: "dependencies",
+          },
+        ];
+
+        const hasOpenAiKey = program.openAiKey || process.env["OPENAI_API_KEY"];
+        if (program.vectorDb === "none" && hasOpenAiKey) {
+          actionChoices.push({
+            title:
+              "Generate code, install dependencies, and run the app (~2 min)",
+            value: "runApp",
+          });
+        }
+
+        const { action } = await prompts(
+          {
+            type: "select",
+            name: "action",
+            message: "How would you like to proceed?",
+            choices: actionChoices,
+            initial: 1,
+          },
+          handlers,
+        );
+
+        program.postInstallAction = action;
+      }
+    }
+  }
+
   if (!program.template) {
     if (ciInfo.isCI) {
       program.template = getPrefOrDefault("template");
@@ -148,6 +192,10 @@ export const askQuestions = async (
               title: `Community template from ${styledRepo}`,
               value: "community",
             },
+            {
+              title: "Example using a LlamaPack",
+              value: "llamapack",
+            },
           ],
           initial: 1,
         },
@@ -181,6 +229,27 @@ export const askQuestions = async (
     return; // early return - no further questions needed for community projects
   }
 
+  if (program.template === "llamapack") {
+    const availableLlamaPacks = await getAvailableLlamapackOptions();
+    const { llamapack } = await prompts(
+      {
+        type: "select",
+        name: "llamapack",
+        message: "Select LlamaPack",
+        choices: availableLlamaPacks.map((pack) => ({
+          title: pack.name,
+          value: pack.folderPath,
+        })),
+        initial: 0,
+      },
+      handlers,
+    );
+    program.llamapack = llamapack;
+    preferences.llamapack = llamapack;
+    await askPostInstallAction();
+    return; // early return - no further questions needed for llamapack projects
+  }
+
   if (!program.framework) {
     if (ciInfo.isCI) {
       program.framework = getPrefOrDefault("framework");
@@ -386,45 +455,7 @@ export const askQuestions = async (
     }
   }
 
-  // Ask for next action after installation
-  if (program.postInstallAction === undefined) {
-    if (ciInfo.isCI) {
-      program.postInstallAction = getPrefOrDefault("postInstallAction");
-    } else {
-      let actionChoices = [
-        {
-          title: "Just generate code (~1 sec)",
-          value: "none",
-        },
-        {
-          title: "Generate code and install dependencies (~2 min)",
-          value: "dependencies",
-        },
-      ];
-
-      const hasOpenAiKey = program.openAiKey || process.env["OPENAI_API_KEY"];
-      if (program.vectorDb === "none" && hasOpenAiKey) {
-        actionChoices.push({
-          title:
-            "Generate code, install dependencies, and run the app (~2 min)",
-          value: "runApp",
-        });
-      }
-
-      const { action } = await prompts(
-        {
-          type: "select",
-          name: "action",
-          message: "How would you like to proceed?",
-          choices: actionChoices,
-          initial: 1,
-        },
-        handlers,
-      );
-
-      program.postInstallAction = action;
-    }
-  }
+  await askPostInstallAction();
 
   // TODO: consider using zod to validate the input (doesn't work like this as not every option is required)
   // templateUISchema.parse(program.ui);

packages/create-llama/templates/components/sample-projects/llamapack/README-template.md

@@ -0,0 +1,16 @@
+---
+
+## Quickstart
+
+1. Check above instructions for setting up your environment and export required environment variables
+   For example, if you are using bash, you can run the following command to set up OpenAI API key
+
+```bash
+export OPENAI_API_KEY=your_api_key
+```
+
+2. Run the example
+
+```
+poetry run python example.py
+```

packages/create-llama/templates/components/sample-projects/llamapack/pyproject.toml

@@ -0,0 +1,16 @@
+[tool.poetry]
+name = "app"
+version = "0.1.0"
+description = "Llama Pack Example"
+authors = ["Marcus Schiesser <mail@marcusschiesser.de>"]
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = "^3.11,<3.12"
+llama-index = "^0.9.19"
+python-dotenv = "^1.0.0"
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"

packages/create-llama/templates/types/simple/express/gitignore

@@ -1,2 +1,3 @@
 # local env files
 .env
+node_modules/

packages/create-llama/templates/types/streaming/express/gitignore

@@ -1,2 +1,3 @@
 # local env files
 .env
+node_modules/