mirror of
https://github.com/run-llama/LlamaIndexTS.git
synced 2026-07-20 02:53:50 -04:00
Compare commits
5 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
 Alex Yang
|
e80029a39a | ||
|
Alex Yang
|
d04cf437f2 | ||
|
Alex Yang
|
2b1b35b897 | ||
|
Marcus Schiesser
|
fa40b36516 | ||
|
yangqiao
|
da8068e9e0 |
@@ -0,0 +1,5 @@
|
||||
---
|
||||
"@llamaindex/cloud": patch
|
||||
---
|
||||
|
||||
chore: bump sdk openapi.json
|
||||
@@ -0,0 +1,5 @@
|
||||
---
|
||||
"@llamaindex/azure": patch
|
||||
---
|
||||
|
||||
Add `fromConnectionString` method to azure storage libs to track the usage vCore.
|
||||
@@ -1,12 +0,0 @@
|
||||
---
|
||||
title: Agents
|
||||
---
|
||||
|
||||
A built-in agent that can take decisions and reasoning based on the tools provided to it.
|
||||
|
||||
## OpenAI Agent
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/agent/openai";
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
@@ -1,28 +0,0 @@
|
||||
---
|
||||
title: Gemini Agent
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSourceGemini from "!raw-loader!../../../../../../../examples/gemini/agent.ts";
|
||||
|
||||
## Installation
|
||||
|
||||
import { Tab, Tabs } from "fumadocs-ui/components/tabs";
|
||||
|
||||
<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
|
||||
```shell tab="npm"
|
||||
npm install llamaindex @llamaindex/google
|
||||
```
|
||||
|
||||
```shell tab="yarn"
|
||||
yarn add llamaindex @llamaindex/google
|
||||
```
|
||||
|
||||
```shell tab="pnpm"
|
||||
pnpm add llamaindex @llamaindex/google
|
||||
```
|
||||
</Tabs>
|
||||
|
||||
## Source
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSourceGemini} />
|
||||
@@ -1,10 +0,0 @@
|
||||
---
|
||||
title: Chat Engine
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/chatEngine";
|
||||
|
||||
Chat Engine is a class that allows you to create a chatbot from a retriever. It is a wrapper around a retriever that allows you to chat with it in a conversational manner.
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
@@ -1,59 +0,0 @@
|
||||
---
|
||||
title: Context-Aware Agent
|
||||
---
|
||||
|
||||
The Context-Aware Agent enhances the capabilities of standard LLM agents by incorporating relevant context from a retriever for each query. This allows the agent to provide more informed and specific responses based on the available information.
|
||||
|
||||
## Usage
|
||||
|
||||
Here's a simple example of how to use the Context-Aware Agent:
|
||||
|
||||
```typescript
|
||||
import {
|
||||
Document,
|
||||
VectorStoreIndex,
|
||||
} from "llamaindex";
|
||||
import { OpenAI, OpenAIContextAwareAgent } from "@llamaindex/openai";
|
||||
|
||||
async function createContextAwareAgent() {
|
||||
// Create and index some documents
|
||||
const documents = [
|
||||
new Document({
|
||||
text: "LlamaIndex is a data framework for LLM applications.",
|
||||
id_: "doc1",
|
||||
}),
|
||||
new Document({
|
||||
text: "The Eiffel Tower is located in Paris, France.",
|
||||
id_: "doc2",
|
||||
}),
|
||||
];
|
||||
|
||||
const index = await VectorStoreIndex.fromDocuments(documents);
|
||||
const retriever = index.asRetriever({ similarityTopK: 1 });
|
||||
|
||||
// Create the Context-Aware Agent
|
||||
const agent = new OpenAIContextAwareAgent({
|
||||
llm: new OpenAI({ model: "gpt-3.5-turbo" }),
|
||||
contextRetriever: retriever,
|
||||
});
|
||||
|
||||
// Use the agent to answer queries
|
||||
const response = await agent.chat({
|
||||
message: "What is LlamaIndex used for?",
|
||||
});
|
||||
|
||||
console.log("Agent Response:", response.response);
|
||||
}
|
||||
|
||||
createContextAwareAgent().catch(console.error);
|
||||
```
|
||||
|
||||
In this example, the Context-Aware Agent uses the retriever to fetch relevant context for each query, allowing it to provide more accurate and informed responses based on the indexed documents.
|
||||
|
||||
## Key Components
|
||||
|
||||
- `contextRetriever`: A retriever (e.g., from a VectorStoreIndex) that fetches relevant documents or passages for each query.
|
||||
|
||||
## Available Context-Aware Agents
|
||||
|
||||
- `OpenAIContextAwareAgent`: A context-aware agent using OpenAI's models.
|
||||
@@ -1,15 +0,0 @@
|
||||
{
|
||||
"title": "Examples",
|
||||
"pages": [
|
||||
"more_examples",
|
||||
"chat_engine",
|
||||
"vector_index",
|
||||
"summary_index",
|
||||
"save_load_index",
|
||||
"context_aware_agent",
|
||||
"agent",
|
||||
"agent_gemini",
|
||||
"local_llm",
|
||||
"other_llms"
|
||||
]
|
||||
}
|
||||
@@ -1,66 +0,0 @@
|
||||
---
|
||||
title: Using other LLM APIs
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/mistral";
|
||||
|
||||
By default LlamaIndex.TS uses OpenAI's LLMs and embedding models, but we support [lots of other LLMs](../modules/llms) including models from Mistral (Mistral, Mixtral), Anthropic (Claude) and Google (Gemini).
|
||||
|
||||
If you don't want to use an API at all you can [run a local model](./local_llm).
|
||||
|
||||
This example runs you through the process of setting up a Mistral model:
|
||||
|
||||
|
||||
## Installation
|
||||
|
||||
import { Tab, Tabs } from "fumadocs-ui/components/tabs";
|
||||
|
||||
<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
|
||||
```shell tab="npm"
|
||||
npm install llamaindex @llamaindex/mistral
|
||||
```
|
||||
|
||||
```shell tab="yarn"
|
||||
yarn add llamaindex @llamaindex/mistral
|
||||
```
|
||||
|
||||
```shell tab="pnpm"
|
||||
pnpm add llamaindex @llamaindex/mistral
|
||||
```
|
||||
</Tabs>
|
||||
|
||||
## Using another LLM
|
||||
|
||||
You can specify what LLM LlamaIndex.TS will use on the `Settings` object, like this:
|
||||
|
||||
```typescript
|
||||
import { MistralAI } from "@llamaindex/mistral";
|
||||
import { Settings } from "llamaindex";
|
||||
|
||||
Settings.llm = new MistralAI({
|
||||
model: "mistral-tiny",
|
||||
apiKey: "<YOUR_API_KEY>",
|
||||
});
|
||||
```
|
||||
|
||||
You can see examples of other APIs we support by checking out "Available LLMs" in the sidebar of our [LLMs section](../modules/llms).
|
||||
|
||||
## Using another embedding model
|
||||
|
||||
A frequent gotcha when trying to use a different API as your LLM is that LlamaIndex will also by default index and embed your data using OpenAI's embeddings. To completely switch away from OpenAI you will need to set your embedding model as well, for example:
|
||||
|
||||
```typescript
|
||||
import { MistralAIEmbedding } from "@llamaindex/mistral";
|
||||
import { Settings } from "llamaindex";
|
||||
|
||||
Settings.embedModel = new MistralAIEmbedding();
|
||||
```
|
||||
|
||||
We support [many different embeddings](../modules/embeddings).
|
||||
|
||||
## Full example
|
||||
|
||||
This example uses Mistral's `mistral-tiny` model as the LLM and Mistral for embeddings as well.
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
@@ -1,8 +0,0 @@
|
||||
---
|
||||
title: Save/Load an Index
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/storageContext";
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
@@ -1,8 +0,0 @@
|
||||
---
|
||||
title: Summary Index
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/summaryIndex";
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
@@ -1,8 +0,0 @@
|
||||
---
|
||||
title: Vector Index
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/vectorIndex";
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
+1
-5
@@ -1,11 +1,7 @@
|
||||
---
|
||||
title: Chatbot tutorial
|
||||
title: Create-Llama
|
||||
---
|
||||
|
||||
Once you've mastered basic [retrieval-augment generation](retrieval_augmented_generation) you may want to create an interface to chat with your data. You can do this step-by-step, but we recommend getting started quickly using `create-llama`.
|
||||
|
||||
## Using create-llama
|
||||
|
||||
`create-llama` is a powerful but easy to use command-line tool that generates a working, full-stack web application that allows you to chat with your data. You can learn more about it on [the `create-llama` README page](https://www.npmjs.com/package/create-llama).
|
||||
|
||||
Run it once and it will ask you a series of questions about the kind of application you want to generate. Then you can customize your application to suit your use-case. To get started, run:
|
||||
+13
-2
@@ -1,10 +1,10 @@
|
||||
---
|
||||
title: See all examples
|
||||
title: Code examples
|
||||
---
|
||||
|
||||
Our GitHub repository has a wealth of examples to explore and try out. You can check out our [examples folder](https://github.com/run-llama/LlamaIndexTS/tree/main/examples) to see them all at once, or browse the pages in this section for some selected highlights.
|
||||
|
||||
## Check out all examples
|
||||
## Use examples locally
|
||||
|
||||
It may be useful to check out all the examples at once so you can try them out locally. To do this into a folder called `my-new-project`, run these commands:
|
||||
|
||||
@@ -19,3 +19,14 @@ Then you can run any example in the folder with `tsx`, e.g.:
|
||||
```bash npm2yarn
|
||||
npx tsx ./vectorIndex.ts
|
||||
```
|
||||
|
||||
## Try examples online
|
||||
|
||||
You can also try the examples online using StackBlitz:
|
||||
|
||||
<iframe
|
||||
className="w-full h-[440px]"
|
||||
aria-label="LlamaIndex.TS Examples"
|
||||
aria-description="This is a list of examples for LlamaIndex.TS."
|
||||
src="https://stackblitz.com/github/run-llama/LlamaIndexTS/tree/main/examples?file=README.md"
|
||||
/>
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Choose Framework
|
||||
title: Frameworks
|
||||
description: We support multiple JS runtime and frameworks, bundlers.
|
||||
---
|
||||
import {
|
||||
@@ -0,0 +1,6 @@
|
||||
{
|
||||
"title": "Framework",
|
||||
"description": "The setup guide",
|
||||
"defaultOpen": true,
|
||||
"pages": ["node", "typescript", "next", "vite", "cloudflare"]
|
||||
}
|
||||
+1
-1
@@ -42,7 +42,7 @@ By the default, we are using `js-tiktoken` for tokenization. You can install `gp
|
||||
```
|
||||
</Tabs>
|
||||
|
||||
> Note: This only works for Node.js
|
||||
**Note**: This only works for Node.js
|
||||
|
||||
## TypeScript support
|
||||
|
||||
|
Before Width: | Height: | Size: 540 KiB After Width: | Height: | Size: 540 KiB |
@@ -37,20 +37,20 @@ In most cases, you'll also need an LLM package to use LlamaIndex. For example, t
|
||||
```
|
||||
</Tabs>
|
||||
|
||||
Go to [Using other LLM APIs](/docs/llamaindex/examples/other_llms) to find out how to use other LLMs.
|
||||
Go to [LLM APIs](/docs/llamaindex/modules/llms) to find out how to use other LLMs.
|
||||
|
||||
|
||||
## What's next?
|
||||
|
||||
<Cards>
|
||||
<Card
|
||||
title="I want to try LlamaIndex.TS"
|
||||
description="Learn how to use LlamaIndex.TS with different JS runtime and frameworks."
|
||||
href="/docs/llamaindex/getting_started/setup"
|
||||
title="Learn LlamaIndex.TS"
|
||||
description="Learn how to use LlamaIndex.TS by starting with one of our tutorials."
|
||||
href="/docs/llamaindex/tutorials/rag"
|
||||
/>
|
||||
<Card
|
||||
title="Show me code examples"
|
||||
description="Explore code examples using LlamaIndex.TS."
|
||||
href="https://stackblitz.com/github/run-llama/LlamaIndexTS/tree/main/examples?file=README.md"
|
||||
href="/docs/llamaindex/getting_started/examples"
|
||||
/>
|
||||
</Cards>
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
{
|
||||
"title": "Getting Started",
|
||||
"pages": ["index", "setup", "starter_tutorial", "environments", "concepts"]
|
||||
"pages": ["index", "create_llama", "examples", "frameworks"]
|
||||
}
|
||||
|
||||
@@ -1,6 +0,0 @@
|
||||
{
|
||||
"title": "Setup",
|
||||
"description": "The setup guide",
|
||||
"defaultOpen": true,
|
||||
"pages": ["index", "next", "node", "typescript", "vite", "cloudflare"]
|
||||
}
|
||||
@@ -1,9 +0,0 @@
|
||||
{
|
||||
"title": "Starter Tutorials",
|
||||
"pages": [
|
||||
"retrieval_augmented_generation",
|
||||
"chatbot",
|
||||
"structured_data_extraction",
|
||||
"agent"
|
||||
]
|
||||
}
|
||||
@@ -1,16 +0,0 @@
|
||||
---
|
||||
title: Cost Analysis
|
||||
---
|
||||
|
||||
This page shows how to track LLM cost using APIs.
|
||||
|
||||
## Callback Manager
|
||||
|
||||
The callback manager is a class that manages the callback functions.
|
||||
|
||||
You can register `llm-start`, `llm-end`, and `llm-stream` callbacks to the callback manager for tracking the cost.
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/recipes/cost-analysis";
|
||||
|
||||
<DynamicCodeBlock lang="ts" code={CodeSource} />
|
||||
@@ -1,5 +0,0 @@
|
||||
{
|
||||
"title": "Guide",
|
||||
"description": "See our guide",
|
||||
"pages": ["loading", "workflow", "chat", "agents", "cost-analysis"]
|
||||
}
|
||||
@@ -16,9 +16,13 @@ The TypeScript implementation is designed for JavaScript server side application
|
||||
|
||||
LlamaIndex.TS provides tools for beginners, advanced users, and everyone in between.
|
||||
|
||||
Try it out with a starter example using StackBlitz:
|
||||
|
||||
<iframe
|
||||
className="w-full h-[440px]"
|
||||
aria-label="LlamaIndex.TS Starter"
|
||||
aria-description="This is a starter example for LlamaIndex.TS, it shows the basic usage of the library."
|
||||
src="https://stackblitz.com/github/run-llama/LlamaIndexTS/tree/main/examples?embed=1&file=starter.ts"
|
||||
/>
|
||||
|
||||
You'll need an OpenAI API key to run this example. You can retrieve it from [OpenAI](https://platform.openai.com/api-keys).
|
||||
@@ -4,13 +4,11 @@
|
||||
"root": true,
|
||||
"pages": [
|
||||
"---Guide---",
|
||||
"what-is-llamaindex",
|
||||
"index",
|
||||
"getting_started",
|
||||
"migration",
|
||||
"guide",
|
||||
"examples",
|
||||
"tutorials",
|
||||
"modules",
|
||||
"integration"
|
||||
"integration",
|
||||
"migration"
|
||||
]
|
||||
}
|
||||
|
||||
+2
@@ -2,6 +2,8 @@
|
||||
title: Agents
|
||||
---
|
||||
|
||||
**Note**: Agents are deprecated, use [Agent Workflows](/docs/llamaindex/modules/agent_workflow) instead.
|
||||
|
||||
An “agent” is an automated reasoning and decision engine. It takes in a user input/query and can make internal decisions for executing that query in order to return the correct result. The key agent components can include, but are not limited to:
|
||||
|
||||
- Breaking down a complex question into smaller ones
|
||||
@@ -1,5 +1,5 @@
|
||||
{
|
||||
"title": "Migration",
|
||||
"description": "Migration between different versions",
|
||||
"pages": ["0.8-to-0.9"]
|
||||
"pages": ["0.8-to-0.9", "deprecated"]
|
||||
}
|
||||
|
||||
@@ -1,9 +1,9 @@
|
||||
---
|
||||
title: Agent Workflow
|
||||
title: Agent Workflows
|
||||
---
|
||||
|
||||
|
||||
Agent Workflows are a powerful system that enables you to create and orchestrate one or multiple agents with tools to perform specific tasks. It's built on top of the base `Workflow` system and provides a streamlined interface for agent interactions.
|
||||
Agent Workflows are a powerful system that enables you to create and orchestrate one or multiple agents with tools to perform specific tasks. It's built on top of the base [`Workflow`](./workflows) system and provides a streamlined interface for agent interactions.
|
||||
|
||||
## Usage
|
||||
|
||||
|
||||
@@ -4,7 +4,7 @@ title: Embedding
|
||||
|
||||
The embedding model in LlamaIndex is responsible for creating numerical representations of text. By default, LlamaIndex will use the `text-embedding-ada-002` model from OpenAI.
|
||||
|
||||
This can be explicitly updated through `Settings`
|
||||
This can be explicitly updated through `Settings.embedModel`.
|
||||
|
||||
## Installation
|
||||
|
||||
@@ -35,7 +35,7 @@ Settings.embedModel = new OpenAIEmbedding({
|
||||
|
||||
## Local Embedding
|
||||
|
||||
For local embeddings, you can use the [HuggingFace](/docs/llamaindex/modules/embeddings/available_embeddings/huggingface) embedding model.
|
||||
For local embeddings, you can use the [HuggingFace](/docs/llamaindex/modules/embeddings/huggingface) embedding model.
|
||||
|
||||
## Local Ollama Embeddings With Remote Host
|
||||
|
||||
|
||||
+1
-1
@@ -3,7 +3,7 @@ title: Groq
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../../../examples/groq.ts";
|
||||
import CodeSource from "!raw-loader!../../../../../../../../examples/groq.ts";
|
||||
|
||||
## Installation
|
||||
|
||||
@@ -45,7 +45,7 @@ export AZURE_OPENAI_DEPLOYMENT="gpt-4" # or some other deployment name
|
||||
|
||||
## Local LLM
|
||||
|
||||
For local LLMs, currently we recommend the use of [Ollama](/docs/llamaindex/modules/llms/available_llms/ollama) LLM.
|
||||
For local LLMs, currently we recommend the use of [Ollama](/docs/llamaindex/modules/llms/ollama) LLM.
|
||||
|
||||
## Available LLMs
|
||||
|
||||
|
||||
+19
-1
@@ -1,8 +1,26 @@
|
||||
---
|
||||
title: Perplexity LLM
|
||||
---
|
||||
## Usage
|
||||
|
||||
## Installation
|
||||
|
||||
import { Tab, Tabs } from "fumadocs-ui/components/tabs";
|
||||
|
||||
<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
|
||||
```shell tab="npm"
|
||||
npm install @llamaindex/perplexity
|
||||
```
|
||||
|
||||
```shell tab="yarn"
|
||||
yarn add @llamaindex/perplexity
|
||||
```
|
||||
|
||||
```shell tab="pnpm"
|
||||
pnpm add @llamaindex/perplexity
|
||||
```
|
||||
</Tabs>
|
||||
|
||||
## Usage
|
||||
|
||||
```ts
|
||||
import { Settings } from "llamaindex";
|
||||
+3
-3
@@ -8,15 +8,15 @@ import { Tab, Tabs } from "fumadocs-ui/components/tabs";
|
||||
|
||||
<Tabs groupId="install" items={["npm", "yarn", "pnpm"]} persist>
|
||||
```shell tab="npm"
|
||||
npm install llamaindex
|
||||
npm install @llamaindex/together
|
||||
```
|
||||
|
||||
```shell tab="yarn"
|
||||
yarn add llamaindex
|
||||
yarn add @llamaindex/together
|
||||
```
|
||||
|
||||
```shell tab="pnpm"
|
||||
pnpm add llamaindex
|
||||
pnpm add @llamaindex/together
|
||||
```
|
||||
</Tabs>
|
||||
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Agent tutorial
|
||||
title: 1. Setup
|
||||
---
|
||||
|
||||
In this guide we'll walk you through the process of building an Agent in JavaScript using the LlamaIndex.TS library, starting from nothing and adding complexity in stages.
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Create a basic agent
|
||||
title: 2. Create a basic agent
|
||||
---
|
||||
|
||||
We want to use `await` so we're going to wrap all of our code in a `main` function, like this:
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Using a local model via Ollama
|
||||
title: 3. Using a local model via Ollama
|
||||
---
|
||||
|
||||
If you're happy using OpenAI, you can skip this section, but many people are interested in using models they run themselves. The easiest way to do this is via the great work of our friends at [Ollama](https://ollama.com/), who provide a simple to use client that will download, install and run a [growing range of models](https://ollama.com/library) for you.
|
||||
+5
-34
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Adding Retrieval-Augmented Generation (RAG)
|
||||
title: 4. Adding Retrieval-Augmented Generation (RAG)
|
||||
---
|
||||
|
||||
While an agent that can perform math is nifty (LLMs are usually not very good at math), LLM-based applications are always more interesting when they work with large amounts of data. In this case, we're going to use a 200-page PDF of the proposed budget of the city of San Francisco for fiscal years 2024-2024 and 2024-2025. It's a great example because it's extremely wordy and full of tables of figures, which present a challenge for humans and LLMs alike.
|
||||
@@ -87,32 +87,9 @@ By default LlamaIndex will retrieve just the 2 most relevant chunks of text. Thi
|
||||
retriever.similarityTopK = 10;
|
||||
```
|
||||
|
||||
### Approach 1: Create a Context-Aware Agent
|
||||
### Use index.queryTool
|
||||
|
||||
With the retriever ready, you can create a **context-aware agent**.
|
||||
|
||||
```javascript
|
||||
const agent = new OpenAIContextAwareAgent({
|
||||
contextRetriever: retriever,
|
||||
});
|
||||
|
||||
// Example query to the context-aware agent
|
||||
let response = await agent.chat({
|
||||
message: `What's the budget of San Francisco in 2023-2024?`,
|
||||
});
|
||||
|
||||
console.log(response);
|
||||
```
|
||||
|
||||
**Expected Output:**
|
||||
|
||||
```md
|
||||
The total budget for the City and County of San Francisco for the fiscal year 2023-2024 is $14.6 billion. This represents a $611.8 million, or 4.4 percent, increase over the previous fiscal year's budget. The budget covers various expenditures across different departments and services, including significant allocations to public works, transportation, commerce, public protection, and health services.
|
||||
```
|
||||
|
||||
### Approach 2: Using QueryEngineTool (Alternative Approach)
|
||||
|
||||
If you prefer more flexibility and don't mind additional complexity, you can create a `QueryEngineTool`. This approach allows you to define the query logic, providing a more tailored way to interact with the data, but note that it introduces a delay due to the extra tool call.
|
||||
`index.queryTool` creates a `QueryEngineTool` that can be used be an agent to query data from the index.
|
||||
|
||||
```javascript
|
||||
const tools = [
|
||||
@@ -125,9 +102,9 @@ const tools = [
|
||||
];
|
||||
|
||||
// Create an agent using the tools array
|
||||
const myAgent = agent({ tools });
|
||||
const ragAgent = agent({ tools });
|
||||
|
||||
let toolResponse = await myAgent.run("What's the budget of San Francisco in 2023-2024?");
|
||||
let toolResponse = await ragAgent.run("What's the budget of San Francisco in 2023-2024?");
|
||||
|
||||
console.log(toolResponse);
|
||||
```
|
||||
@@ -155,10 +132,4 @@ console.log(toolResponse);
|
||||
|
||||
Once again we see a `toolResult`. You can see the query the LLM decided to send to the query engine ("total budget"), and the output the engine returned. In `response.message` you see that the LLM has returned the output from the tool almost verbatim, although it trimmed out the bit about 2024-2025 since we didn't ask about that year.
|
||||
|
||||
### Comparison of Approaches
|
||||
|
||||
The `OpenAIContextAwareAgent` approach simplifies the setup by allowing you to directly link the retriever to the agent, making it straightforward to access relevant context for your queries. This is ideal for situations where you want easy integration with existing data sources, like a context chat engine.
|
||||
|
||||
On the other hand, using the `QueryEngineTool` offers more flexibility and power. This method allows for customization in how queries are constructed and executed, enabling you to query data from various storages and process them in different ways. However, this added flexibility comes with increased complexity and response time due to the separate tool call and queryEngine generating tool output by LLM that is then passed to the agent.
|
||||
|
||||
So now we have an agent that can index complicated documents and answer questions about them. Let's [combine our math agent and our RAG agent](5_rag_and_tools)!
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: A RAG agent that does math
|
||||
title: 5. A RAG agent that does math
|
||||
---
|
||||
|
||||
In [our third iteration of the agent](https://github.com/run-llama/ts-agents/blob/main/3_rag_and_tools/agent.ts) we've combined the two previous agents, so we've defined both `sumNumbers` and a `QueryEngineTool` and created an array of two tools. The tools support both Zod and JSON Schema for parameter definition:
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Adding LlamaParse
|
||||
title: 6. Adding LlamaParse
|
||||
---
|
||||
|
||||
Complicated PDFs can be very tricky for LLMs to understand. To help with this, LlamaIndex provides LlamaParse, a hosted service that parses complex documents including PDFs. To use it, get a `LLAMA_CLOUD_API_KEY` by [signing up for LlamaCloud](https://cloud.llamaindex.ai/) (it's free for up to 1000 pages/day) and adding it to your `.env` file just as you did for your OpenAI key:
|
||||
+1
-1
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Adding persistent vector storage
|
||||
title: 7. Adding persistent vector storage
|
||||
---
|
||||
|
||||
In the previous examples, we've been loading our data into memory each time we run the agent. This is fine for small datasets, but for larger datasets you'll want to store your embeddings in a database. LlamaIndex.TS provides a `VectorStore` class that can store your embeddings in a variety of databases. We're going to use [Qdrant](https://qdrant.tech/), a popular vector store, for this example.
|
||||
|
Before Width: | Height: | Size: 236 KiB After Width: | Height: | Size: 236 KiB |
+1
-1
@@ -1,5 +1,5 @@
|
||||
{
|
||||
"title": "Agents",
|
||||
"title": "Agent with RAG",
|
||||
"pages": [
|
||||
"1_setup",
|
||||
"2_create_agent",
|
||||
+3
-3
@@ -1,11 +1,11 @@
|
||||
---
|
||||
title: Agent tutorial
|
||||
title: Basic Agent
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../../examples/agent/openai";
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/agent/openai";
|
||||
|
||||
We have a comprehensive, step-by-step [guide to building agents in LlamaIndex.TS](../../guides/agents/setup) that we recommend to learn what agents are and how to build them for production. But building a basic agent is simple:
|
||||
We have a comprehensive, step-by-step [guide to building agents in LlamaIndex.TS](./agents/1_setup) that we recommend to learn what agents are and how to build them for production. But building a basic agent is simple:
|
||||
|
||||
## Set up
|
||||
|
||||
+5
-2
@@ -4,7 +4,7 @@ title: Local LLMs
|
||||
|
||||
import { Tab, Tabs } from "fumadocs-ui/components/tabs";
|
||||
|
||||
LlamaIndex.TS supports OpenAI and [other remote LLM APIs](other_llms). You can also run a local LLM on your machine!
|
||||
LlamaIndex.TS supports OpenAI and [other remote LLM APIs](/docs/llamaindex/modules/llms). You can also run a local LLM on your machine!
|
||||
|
||||
## Using a local model via Ollama
|
||||
|
||||
@@ -45,7 +45,10 @@ To switch the LLM in your code, you first need to make sure to install the packa
|
||||
Then, to tell LlamaIndex to use a local LLM, use the `Settings` object:
|
||||
|
||||
```javascript
|
||||
Settings.llm = new Ollama({
|
||||
import { Settings } from "llamaindex";
|
||||
import { ollama } from "@llamaindex/ollama";
|
||||
|
||||
Settings.llm = ollama({
|
||||
model: "mixtral:8x7b",
|
||||
});
|
||||
```
|
||||
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"title": "Tutorials",
|
||||
"pages": [
|
||||
"rag",
|
||||
"basic_agent",
|
||||
"agents",
|
||||
"workflow",
|
||||
"local_llm",
|
||||
"chatbot",
|
||||
"structured_data_extraction"
|
||||
]
|
||||
}
|
||||
|
Before Width: | Height: | Size: 27 KiB After Width: | Height: | Size: 27 KiB |
|
Before Width: | Height: | Size: 49 KiB After Width: | Height: | Size: 49 KiB |
|
Before Width: | Height: | Size: 36 KiB After Width: | Height: | Size: 36 KiB |
+3
-3
@@ -16,7 +16,7 @@ LlamaIndex uses a two stage method when using an LLM with your data:
|
||||
1. **indexing stage**: preparing a knowledge base, and
|
||||
2. **querying stage**: retrieving relevant context from the knowledge to assist the LLM in responding to a question
|
||||
|
||||
|
||||
|
||||
|
||||
This process is also known as Retrieval Augmented Generation (RAG).
|
||||
|
||||
@@ -28,7 +28,7 @@ Let's explore each stage in detail.
|
||||
|
||||
LlamaIndex.TS help you prepare the knowledge base with a suite of data connectors and indexes.
|
||||
|
||||
|
||||
|
||||
|
||||
[**Data Loaders**](/docs/llamaindex/modules/data_loaders/index):
|
||||
A data connector (i.e. `Reader`) ingest data from different data sources and data formats into a simple `Document` representation (text and simple metadata).
|
||||
@@ -54,7 +54,7 @@ LlamaIndex provides composable modules that help you build and integrate RAG pip
|
||||
|
||||
These building blocks can be customized to reflect ranking preferences, as well as composed to reason over multiple knowledge bases in a structured way.
|
||||
|
||||
|
||||
|
||||
|
||||
#### Building Blocks
|
||||
|
||||
+3
-3
@@ -1,12 +1,12 @@
|
||||
---
|
||||
title: Retrieval Augmented Generation (RAG) Tutorial
|
||||
title: Retrieval Augmented Generation (RAG)
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../../examples/vectorIndex";
|
||||
import TSConfigSource from "!!raw-loader!../../../../../../../../examples/tsconfig.json";
|
||||
|
||||
One of the most common use-cases for LlamaIndex is Retrieval-Augmented Generation or RAG, in which your data is indexed and selectively retrieved to be given to an LLM as source material for responding to a query. You can learn more about the [concepts behind RAG](../concepts).
|
||||
One of the most common use-cases for LlamaIndex is Retrieval-Augmented Generation or RAG, in which your data is indexed and selectively retrieved to be given to an LLM as source material for responding to a query. You can learn more about the [concepts behind RAG](./rag/concepts).
|
||||
|
||||
## Set up the project
|
||||
|
||||
@@ -19,7 +19,7 @@ npm install -D typescript @types/node
|
||||
|
||||
Then, check out the [installation](../setup) steps to install LlamaIndex.TS and prepare an OpenAI key.
|
||||
|
||||
You can use [other LLMs](../../examples/other_llms) via their APIs; if you would prefer to use local models check out our [local LLM example](../../examples/local_llm).
|
||||
You can use [other LLMs](/docs/llamaindex/modules/llms) via their APIs; if you would prefer to use local models check out our [local LLM example](./local_llm).
|
||||
|
||||
## Run queries
|
||||
|
||||
+3
-3
@@ -1,13 +1,13 @@
|
||||
---
|
||||
title: Structured data extraction tutorial
|
||||
title: Structured data extraction
|
||||
---
|
||||
|
||||
import { DynamicCodeBlock } from 'fumadocs-ui/components/dynamic-codeblock';
|
||||
import CodeSource from "!raw-loader!../../../../../../../../examples/jsonExtract";
|
||||
import CodeSource from "!raw-loader!../../../../../../../examples/jsonExtract";
|
||||
|
||||
Make sure you have installed LlamaIndex.TS and have an OpenAI key. If you haven't, check out the [installation](../setup) guide.
|
||||
|
||||
You can use [other LLMs](../../examples/other_llms) via their APIs; if you would prefer to use local models check out our [local LLM example](../../examples/local_llm).
|
||||
You can use [other LLMs](/docs/llamaindex/modules/llms) via their APIs; if you would prefer to use local models check out our [local LLM example](./local_llm).
|
||||
|
||||
## Set up
|
||||
|
||||
@@ -1,36 +1,31 @@
|
||||
/**
|
||||
* This example shows how to use AgentWorkflow as a single agent with tools
|
||||
* This example shows how to use a single agent with a tool
|
||||
*/
|
||||
import { openai } from "@llamaindex/openai";
|
||||
import { Settings, agent } from "llamaindex";
|
||||
import { agent } from "llamaindex";
|
||||
import { getWeatherTool } from "../agent/utils/tools";
|
||||
|
||||
Settings.llm = openai({
|
||||
model: "gpt-4o",
|
||||
});
|
||||
|
||||
async function singleWeatherAgent() {
|
||||
const workflow = agent({
|
||||
async function main() {
|
||||
const weatherAgent = agent({
|
||||
llm: openai({
|
||||
model: "gpt-4o",
|
||||
}),
|
||||
tools: [getWeatherTool],
|
||||
verbose: false,
|
||||
});
|
||||
|
||||
const workflowContext = workflow.run(
|
||||
"What's the weather like in San Francisco?",
|
||||
);
|
||||
const sfResult = await workflowContext;
|
||||
// The weather in San Francisco, CA is currently sunny.
|
||||
console.log(`${JSON.stringify(sfResult, null, 2)}`);
|
||||
// Run the agent and keep the context
|
||||
const context = weatherAgent.run("What's the weather like in San Francisco?");
|
||||
const result = await context;
|
||||
console.log(`${JSON.stringify(result, null, 2)}`);
|
||||
|
||||
// Reuse the context from the previous run
|
||||
const workflowContext2 = workflow.run("Compare it with California?", {
|
||||
context: workflowContext.data,
|
||||
const caResult = await weatherAgent.run("Compare it with California?", {
|
||||
context: context.data,
|
||||
});
|
||||
const caResult = await workflowContext2;
|
||||
// Both San Francisco and California are currently experiencing sunny weather.
|
||||
console.log(`${JSON.stringify(caResult, null, 2)}`);
|
||||
}
|
||||
|
||||
singleWeatherAgent().catch((error) => {
|
||||
main().catch((error) => {
|
||||
console.error("Error:", error);
|
||||
});
|
||||
|
||||
+2165
-482
File diff suppressed because it is too large
Load Diff
@@ -147,6 +147,8 @@ export class LlamaParseReader extends FileReader {
|
||||
ignore_document_elements_for_layout_detection?: boolean | undefined;
|
||||
complemental_formatting_instruction?: string | undefined;
|
||||
content_guideline_instruction?: string | undefined;
|
||||
adaptive_long_table?: boolean | undefined;
|
||||
model?: string | undefined;
|
||||
|
||||
constructor(
|
||||
params: Partial<Omit<LlamaParseReader, "language" | "apiKey">> & {
|
||||
@@ -296,6 +298,8 @@ export class LlamaParseReader extends FileReader {
|
||||
complemental_formatting_instruction:
|
||||
this.complemental_formatting_instruction,
|
||||
content_guideline_instruction: this.content_guideline_instruction,
|
||||
adaptive_long_table: this.adaptive_long_table,
|
||||
model: this.model,
|
||||
} satisfies {
|
||||
[Key in keyof Body_upload_file_api_v1_parsing_upload_post]-?:
|
||||
| Body_upload_file_api_v1_parsing_upload_post[Key]
|
||||
|
||||
@@ -5,8 +5,8 @@
|
||||
## Current Features:
|
||||
|
||||
- Bedrock support for Amazon Nova models Pro, Lite and Micro
|
||||
- Bedrock support for the Anthropic Claude Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock) including the latest Sonnet 3.5 v2 and Haiku 3.5
|
||||
- Bedrock support for the Meta LLama 2, 3, 3.1 and 3.2 Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
|
||||
- Bedrock support for the Anthropic Claude Models [usage](https://ts.llamaindex.ai/docs/llamaindex/modules/llms/bedrock) including the latest Sonnet 3.5 v2 and Haiku 3.5
|
||||
- Bedrock support for the Meta LLama 2, 3, 3.1 and 3.2 Models [usage](https://ts.llamaindex.ai/docs/llamaindex/modules/llms/bedrock)
|
||||
- Meta LLama3.1 405b and Llama3.2 tool call support
|
||||
- Meta 3.2 11B and 90B vision support
|
||||
- Bedrock support for querying Knowledge Base
|
||||
|
||||
@@ -57,6 +57,21 @@ export class AzureCosmosVCoreChatStore<
|
||||
});
|
||||
}
|
||||
|
||||
static fromConnectionString(
|
||||
connectionString: string,
|
||||
dbName: string = DEFAULT_CHAT_DATABASE,
|
||||
collectionName: string = DEFAULT_CHAT_Collection,
|
||||
): AzureCosmosVCoreChatStore {
|
||||
const mongoClient = new MongoClient(connectionString, {
|
||||
appName: "LLAMAINDEX_JS",
|
||||
});
|
||||
return new AzureCosmosVCoreChatStore({
|
||||
mongoClient,
|
||||
dbName,
|
||||
collectionName,
|
||||
});
|
||||
}
|
||||
|
||||
client(): MongoClient {
|
||||
return this.mongoClient;
|
||||
}
|
||||
|
||||
@@ -46,4 +46,33 @@ export class AzureCosmosVCoreDocumentStore extends KVDocumentStore {
|
||||
namespace,
|
||||
});
|
||||
}
|
||||
|
||||
/**
|
||||
* Static method for creating an instance using a connection string.
|
||||
* @returns Instance of AzureCosmosVCoreDocumentStore
|
||||
* @param connectionString - MongoDB connection string
|
||||
* @param dbName - Database name
|
||||
* @param collectionName - Collection name
|
||||
* @example
|
||||
* ```ts
|
||||
* const indexStore = AzureCosmosVCoreDocumentStore.fromConnectionString("mongodb://localhost:27017", "my_db", "my_collection");
|
||||
* ```
|
||||
*/
|
||||
static fromConnectionString(
|
||||
connectionString: string,
|
||||
dbName: string = DEFAULT_DATABASE,
|
||||
collectionName: string = DEFAULT_COLLECTION,
|
||||
): AzureCosmosVCoreDocumentStore {
|
||||
const mongoClient = new MongoClient(connectionString, {
|
||||
appName: "LLAMAINDEX_JS",
|
||||
});
|
||||
return new AzureCosmosVCoreDocumentStore({
|
||||
azureCosmosVCoreKVStore: new AzureCosmosVCoreKVStore({
|
||||
mongoClient,
|
||||
dbName,
|
||||
collectionName,
|
||||
}),
|
||||
namespace: `${dbName}.${collectionName}`,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
@@ -46,4 +46,33 @@ export class AzureCosmosVCoreIndexStore extends KVIndexStore {
|
||||
namespace,
|
||||
});
|
||||
}
|
||||
|
||||
/**
|
||||
* Static method for creating an instance using a connection string.
|
||||
* @returns Instance of AzureCosmosVCoreIndexStore
|
||||
* @param connectionString - MongoDB connection string
|
||||
* @param dbName - Database name
|
||||
* @param collectionName - Collection name
|
||||
* @example
|
||||
* ```ts
|
||||
* const indexStore = AzureCosmosVCoreIndexStore.fromConnectionString("mongodb://localhost:27017", "my_db", "my_collection");
|
||||
* ```
|
||||
*/
|
||||
static fromConnectionString(
|
||||
connectionString: string,
|
||||
dbName: string = DEFAULT_DATABASE,
|
||||
collectionName: string = DEFAULT_COLLECTION,
|
||||
): AzureCosmosVCoreIndexStore {
|
||||
const mongoClient = new MongoClient(connectionString, {
|
||||
appName: "LLAMAINDEX_JS",
|
||||
});
|
||||
return new AzureCosmosVCoreIndexStore({
|
||||
azureCosmosVCoreKVStore: new AzureCosmosVCoreKVStore({
|
||||
mongoClient,
|
||||
dbName,
|
||||
collectionName,
|
||||
}),
|
||||
namespace: `${dbName}.${collectionName}`,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user