### Traditional Multi-Provider Setup Source: https://developer.puter.com/tutorials/access-any-model-using-langchain Shows the standard approach of installing separate packages for each LLM provider. ```bash # The traditional way — one package per provider pip install langchain-openai # for GPT models pip install langchain-anthropic # for Claude models pip install langchain-google-genai # for Gemini models ``` -------------------------------- ### Implement Peer-to-Peer Chat with Puter.js Source: https://developer.puter.com/blog/puter-peer A complete HTML/JavaScript example demonstrating how to start a peer server and connect to it using invite codes. ```html

Peer Chat

Open this page in two tabs. Start a server in one tab, then connect from the other.



    


```

--------------------------------

### Navigate and Install Dependencies

Source: https://developer.puter.com/tutorials/puter-js-react

Change directory into the new project and install its dependencies.

```bash
cd puter-react
npm install
```

--------------------------------

### Handle HTTP Methods and Headers with puter.net.fetch()

Source: https://developer.puter.com/tutorials/cors-free-fetch-api

This example shows how to perform GET, POST, PUT, and DELETE requests using puter.net.fetch(). The syntax mirrors the native fetch API. Ensure the puter.js library is included.

```html


    

        
        
        
        
    

    

    
    
    



```

--------------------------------

### Initialize Project and Install Dependencies

Source: https://developer.puter.com/tutorials/access-any-model-using-langchain

Set up a new Python project and install the required langchain-openai package using uv.

```bash
uv init puter-langchain
cd puter-langchain
uv add langchain-openai
```

--------------------------------

### Basic Chat Completion Example

Source: https://developer.puter.com/tutorials/use-litellm-with-puter

A simple example to get a single chat completion response from the model and print its content. The response format mirrors OpenAI's.

```python
from litellm import completion

response = completion(
    model="openai/gpt-5-nano",
    api_base="https://api.puter.com/puterai/openai/v1/",
    api_key="YOUR_PUTER_AUTH_TOKEN",
    messages=[
        {"role": "user", "content": "What is the capital of France?"},
    ],
)

print(response.choices[0].message.content)
```

--------------------------------

### Deploy Simple 'Hello World' Website with Puter.js

Source: https://developer.puter.com/tutorials/free-unlimited-hosting-api

Use `puter.fs.write()` to create an HTML file and `puter.hosting.create()` to deploy it. This example demonstrates creating a directory, writing an index.html file, and deploying the site with a random subdomain.

```javascript
// Create a directory to write the HTML file
await puter.fs.mkdir('my-website');

// Write an index.html file
await puter.fs.write('my-website/index.html', '
Hello World!
');

// Deploy the website
const site = await puter.hosting.create('hello-world', 'my-website');

console.log(`Website deployed at: https://${site.subdomain}.puter.site`);

```

```html


    
    



```

--------------------------------

### Generate Image with FLUX.1 [schnell] (npm)

Source: https://developer.puter.com/ai/black-forest-labs/flux-schnell

Use the Puter.js library in your Node.js or frontend project to generate an image. Ensure you have installed the library using `npm install @heyputer/puter.js`. This example appends the generated image to the document body.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.txt2img("A serene mountain landscape at sunset", {
    model: "black-forest-labs/flux-schnell"
}).then(image => {
    document.body.appendChild(image);
});

```

--------------------------------

### Retrieve website information using puter.hosting.get

Source: https://developer.puter.com/tutorials/free-unlimited-hosting-api

Demonstrates creating a website and subsequently fetching its metadata including subdomain, URL, and root directory path.

```html


    
    


```

--------------------------------

### Start Development Server

Source: https://developer.puter.com/tutorials/puter-js-next-js

Command to launch the local development server.

```bash
npm run dev
```

--------------------------------

### Generate Image with Grok 2 (Node.js)

Source: https://developer.puter.com/ai/x-ai/grok-2-image

Use the Puter.js AI API to generate an image from a text prompt using the Grok 2 Image model. No API keys or setup are required. Install the library using `npm install @heyputer/puter.js`.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.txt2img("A serene mountain landscape at sunset", {
    model: "x-ai/grok-2-image"
}).then(image => {
    document.body.appendChild(image);
});

```

--------------------------------

### Create a POST Endpoint

Source: https://developer.puter.com/tutorials/serverless-functions-on-puter

Use the router API to create a POST endpoint that returns a JSON response. This is a basic example for starting with serverless functions.

```javascript
router.post("/api/hello", async ({ request, user, params }) => {
  return { message: "Hello, World!" };
});
```

--------------------------------

### Full Example: ElevenLabs Text-to-Speech with Format Selection

Source: https://developer.puter.com/tutorials/free-unlimited-elevenlabs-api

This HTML example integrates the Puter SDK to allow users to input text, select an audio output format from a dropdown, and generate speech using the ElevenLabs provider.

```html


    
    

    
    

    
    



```

--------------------------------

### Manage TCP Socket Events and Data

Source: https://developer.puter.com/networking

Handle the 'open' event for a TCP socket to send data. This example demonstrates sending an HTTP GET request.

```javascript
// Open a TCP socket connection
const socket = new puter.net.Socket("example.com", 80);

// Open a secure TLS socket connection
const tlsSocket = new puter.net.tls.TLSSocket("example.com", 443);

// Send data when connected
socket.on("open", () => {
    socket.write("GET / HTTP/1.1\r\nHost: example.com\r\n\r\n");
});

// Fetch without CORS restrictions
const response = await puter.net.fetch("https://api.example.com");

```

--------------------------------

### Generate Image with GPT Image 2 (Node.js)

Source: https://developer.puter.com/ai/openai/gpt-image-2

Use the Puter.js AI API to generate an image from a text prompt. This example requires the `@heyputer/puter.js` package to be installed.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.txt2img("A serene mountain landscape at sunset", {
    model: "openai/gpt-image-2"
}).then(image => {
    document.body.appendChild(image);
});

```

--------------------------------

### Installation Methods

Source: https://developer.puter.com/blog/aion-2-in-puter-js

Provides options for integrating Puter.js via npm or a direct script tag.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';
```

```html

```

--------------------------------

### Chat with Z.AI using Puter.js (npm)

Source: https://developer.puter.com/ai/z-ai

Use this snippet to interact with Z.AI models via npm. Ensure you have installed the Puter.js library. This example uses the 'z-ai/glm-5' model.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "z-ai/glm-5"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### Tool/Function Calling Setup

Source: https://developer.puter.com/tutorials/use-litellm-with-puter

Prepare for tool/function calling by importing necessary libraries like `json` and `litellm`.

```python
import json
from litellm import completion
```

--------------------------------

### Initialize Node.js Project

Source: https://developer.puter.com/tutorials/puter-js-node-js

Create a new directory, initialize a Node.js project, and enable ES modules.

```bash
mkdir puter-node
cd puter-node
npm init -y

```

```json
{
  "name": "puter-node",
  "type": "module"
}

```

```bash
npm i @heyputer/puter.js

```

--------------------------------

### Chat with Qwen using Puter.js (Node.js)

Source: https://developer.puter.com/ai/qwen

Install the Puter.js library via npm and use the `puter.ai.chat` method to interact with Qwen models. This example uses the `qwen/qwen3.5-flash-02-23` model.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "qwen/qwen3.5-flash-02-23"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### Deploy a multi-page website using Puter SDK

Source: https://developer.puter.com/tutorials/free-unlimited-hosting-api

Creates a directory, populates it with site assets, and deploys the content to a unique subdomain.

```html



    
    


```

--------------------------------

### Chat with Moonshot v1 Auto using Python (OpenAI client)

Source: https://developer.puter.com/ai/moonshotai/moonshot-v1-auto

Interact with Moonshot v1 Auto using the OpenAI Python client. Ensure you have installed the library with `pip install openai`. Configure the `base_url` to `https://api.puter.com/puterai/openai/v1/` and use your PUTER_AUTH_TOKEN for authentication. This example prints the content of the first message choice.

```python
# pip install openai
from openai import OpenAI

client = OpenAI(
    base_url="https://api.puter.com/puterai/openai/v1/",
    api_key="YOUR_PUTER_AUTH_TOKEN",
)

response = client.chat.completions.create(
    model="moonshotai/moonshot-v1-auto",
    messages=[
        {"role": "user", "content": "Explain quantum computing in simple terms"}
    ],
)

print(response.choices[0].message.content)

```

--------------------------------

### Chat with Perceptron Mk1 using Puter.js (npm)

Source: https://developer.puter.com/ai/perceptron

Install Puter.js via npm and use the `puter.ai.chat` method to interact with the Perceptron Mk1 model. This example is suitable for Node.js environments.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "perceptron/perceptron-mk1"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### Create Directory, Write, and List Contents

Source: https://developer.puter.com/tutorials/free-unlimited-cloud-storage-api

Shows how to create a new directory, write a file into it, and then list the contents of that directory. This is useful for organizing files.

```html


    
    


```

--------------------------------

### Basic Text Generation with DeepSeek v4 Flash

Source: https://developer.puter.com/tutorials/free-unlimited-deepseek-api

Use DeepSeek v4 Flash for everyday text generation tasks. This example shows how to get a simple explanation of quantum entanglement.

```html


    
    


```

--------------------------------

### Initialize Next.js Project

Source: https://developer.puter.com/tutorials/puter-js-next-js

Commands to create a new Next.js project and navigate into the directory.

```bash
npx create-next-app@latest puter-nextjs
```

```bash
cd puter-nextjs
```

--------------------------------

### Full HTML Example for HiDream I1 Full

Source: https://developer.puter.com/tutorials/free-unlimited-hidream-api

A complete HTML file demonstrating how to integrate HiDream I1 Full image generation using Puter.js.

```html


    
HiDream I1 Full Image Generation

    
    



```

--------------------------------

### Get File Information with Puter.js

Source: https://developer.puter.com/tutorials/free-unlimited-cloud-storage-api

Retrieve detailed information about files and directories, including name, size, path, creation, and modification times. This example first writes a file to ensure it exists.

```html


    
    



```

--------------------------------

### Generate Image from Text (npm)

Source: https://developer.puter.com/ai/stabilityai

Use the Puter.js library to generate an image from a text prompt. Ensure you have installed the library using npm. This example appends the generated image element to the document body.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.txt2img("A beautiful sunset", {
    model: "stabilityai/stable-diffusion-3-medium"
}).then(imageElement => {
    document.body.appendChild(imageElement);
});

```

--------------------------------

### Python (OpenAI SDK)

Source: https://developer.puter.com/ai/qwen/qwen3-14b

Example of how to use Qwen3 14B with the OpenAI Python client library, configured for the Puter.com API.

```APIDOC
## Chat with Qwen3 14B using Python and OpenAI SDK

### Description
This snippet demonstrates using the `openai` Python library to interact with the Qwen3 14B model via the Puter.com API endpoint.

### Method
`client.chat.completions.create(model, messages)`

### Parameters
- **model** (string) - Required - The identifier for the AI model, e.g., "qwen/qwen3-14b".
- **messages** (array) - Required - A list of message objects, each with a `role` and `content`.
  - **role** (string) - The role of the message sender (e.g., "user", "system", "assistant").
  - **content** (string) - The text of the message.

### Request Example
```python
# pip install openai
from openai import OpenAI

client = OpenAI(
    base_url="https://api.puter.com/puterai/openai/v1/",
    api_key="YOUR_PUTER_AUTH_TOKEN",
)

response = client.chat.completions.create(
    model="qwen/qwen3-14b",
    messages=[
        {"role": "user", "content": "Explain quantum computing in simple terms"}
    ],
)

print(response.choices[0].message.content)
```
```

--------------------------------

### Full HTML Example for Code Generation

Source: https://developer.puter.com/tutorials/free-unlimited-poolside-ai-api

A complete HTML file demonstrating code generation using Laguna M.1 via Puter.js.

```html


    
    


```

--------------------------------

### Chat with Moonshot v1 32K Vision using Python

Source: https://developer.puter.com/ai/moonshotai/moonshot-v1-32k-vision-preview

This Python example uses the OpenAI client library to communicate with the model. Ensure you have the 'openai' library installed and replace 'YOUR_PUTER_AUTH_TOKEN' with your actual token.

```python
# pip install openai
from openai import OpenAI

client = OpenAI(
    base_url="https://api.puter.com/puterai/openai/v1/",
    api_key="YOUR_PUTER_AUTH_TOKEN",
)

response = client.chat.completions.create(
    model="moonshotai/moonshot-v1-32k-vision-preview",
    messages=[
        {"role": "user", "content": "Explain quantum computing in simple terms"}
    ],
)

print(response.choices[0].message.content)

```

--------------------------------

### Chat with Morph AI using Puter.js (npm)

Source: https://developer.puter.com/ai/morph

Use this snippet to initiate a chat with a Morph AI model via npm. Ensure you have installed the @heyputer/puter.js package. This example uses the 'morph/morph-v3-fast' model.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "morph/morph-v3-fast"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### Traditional LLM API Key Setup

Source: https://developer.puter.com/tutorials/litellm-with-openai-claude-gemini-grok

Demonstrates the traditional method of setting environment variables for different LLM providers. This highlights the complexity LiteLLM with Puter aims to simplify.

```bash
# The traditional way — one API key per provider
export OPENAI_API_KEY="sk-..."       # for GPT models
export ANTHROPIC_API_KEY="sk-ant-..."  # for Claude models
export GEMINI_API_KEY="AIza..."       # for Gemini models
```

--------------------------------

### Chat with Meituan AI using Puter.js (npm)

Source: https://developer.puter.com/ai/meituan

Use this snippet to interact with Meituan AI models via npm. Ensure you have installed the @heyputer/puter.js package. This example demonstrates a basic chat request.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "meituan/longcat-flash-chat"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### Full Text-to-Video Generation Example

Source: https://developer.puter.com/tutorials/free-unlimited-wan-ai-api

A complete HTML example demonstrating WAN 2.2 Text-to-Video generation. It includes a status indicator to show loading progress and finalization time. Error handling is also included.

```html


    
Wan 2.2 Text-to-Video Generation

    

    
    


```

--------------------------------

### Chat with Gryphe AI using Puter.js (npm)

Source: https://developer.puter.com/ai/gryphe

Use this snippet to interact with Gryphe AI models via npm. Ensure you have installed the Puter.js library. This example uses the MythoMax 13B model.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "gryphe/mythomax-l2-13b"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### JavaScript (npm)

Source: https://developer.puter.com/ai/openai/gpt-audio-mini

Example of using GPT Audio Mini with Puter.js in a Node.js environment.

```APIDOC
## JavaScript (npm)

### Description
Use the Puter.js library to interact with GPT Audio Mini in a Node.js application.

### Method
`puter.ai.chat(prompt, options)`

### Parameters
- **prompt** (string) - Required - The text prompt for the AI.
- **options** (object) - Required - Configuration options for the AI call.
  - **model** (string) - Required - Specifies the model to use, e.g., "openai/gpt-audio-mini".

### Request Example
```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain quantum computing in simple terms", {
    model: "openai/gpt-audio-mini"
}).then(response => {
    console.log(response.message.content);
});
```

### Response
#### Success Response
- **message.content** (string) - The AI's response content.
```

--------------------------------

### Complete Example: Cohere Text Generation

Source: https://developer.puter.com/tutorials/free-unlimited-cohere-api

A full HTML example demonstrating how to include the Puter.js library and use it to generate text with the Cohere Command model.

```html


    
    


```

--------------------------------

### Chat with Essential AI using Puter.js (npm)

Source: https://developer.puter.com/ai/essentialai

Use this snippet to interact with Essential AI models via npm. Ensure you have installed the Puter.js library. This example demonstrates a basic chat interaction.

```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain AI like I'm five!", {
    model: "essentialai/rnj-1-instruct"
}).then(response => {
    console.log(response);
});

```

--------------------------------

### Make a Text-to-Speech API Call with ElevenLabs

Source: https://developer.puter.com/tutorials/how-to-get-elevenlabs-api-key

Use the ElevenLabs SDK to convert text to speech. This example shows how to get the raw response, including headers like character count and request ID.

```javascript
import { ElevenLabsClient } from '@elevenlabs/elevenlabs-js';

const client = new ElevenLabsClient({ apiKey: 'your_api_key' });

// Get raw response with headers
const { data, rawResponse } = await client.textToSpeech
  .convert('voice_id', {
    text: 'Hello, world!',
    modelId: 'eleven_multilingual_v2',
  })
  .withRawResponse();

// Access character cost from headers
const charCost = rawResponse.headers.get('x-character-count');
const requestId = rawResponse.headers.get('request-id');
const audioData = data;

```

--------------------------------

### JavaScript (npm)

Source: https://developer.puter.com/ai/qwen/qwen3-14b

Example of how to use Qwen3 14B with Puter.js in a Node.js environment.

```APIDOC
## Chat with Qwen3 14B using Puter.js

### Description
This snippet demonstrates how to send a chat message to the Qwen3 14B model using the Puter.js library.

### Method
`puter.ai.chat(prompt, options)`

### Parameters
- **prompt** (string) - Required - The user's message or question.
- **options** (object) - Required - Configuration options for the AI call.
  - **model** (string) - Required - The identifier for the AI model to use, e.g., "qwen/qwen3-14b".

### Response
- **response.message.content** (string) - The content of the AI's response.

### Request Example
```javascript
// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';

puter.ai.chat("Explain quantum computing in simple terms", {
    model: "qwen/qwen3-14b"
}).then(response => {
    document.body.innerHTML = response.message.content;
});
```
```