venice-dev-tools

Venice Dev Tools Documentation

Welcome to the Venice Dev Tools documentation. This SDK provides a simple, intuitive interface for interacting with the Venice AI API.

GitHub Repository Venice AI Platform

Getting Started

The Venice Dev Tools allows you to interact with the Venice AI API, which provides access to powerful AI models for text generation, image creation, and more.

Installation

Install the Venice Dev Tools SDK using npm:

# Install the complete SDK (includes all packages)
npm install venice-dev-tools

# Or install individual packages
# Node.js only
npm install @venice-dev-tools/node

# Browser only
npm install @venice-dev-tools/web

# Core functionality only
npm install @venice-dev-tools/core

Or install globally to use the CLI:

npm install -g @venice-dev-tools/node

Authentication

The Venice Dev Tools requires an API key for authentication. You can obtain an API key from the Venice AI website.

const venice = new VeniceNode({
  apiKey: 'your-api-key',
});

For CLI usage, configure your API key:

venice configure

Basic Usage

Here’s a simple example to get you started:

import { VeniceNode } from '@venice-dev-tools/node';

// Initialize the client
const venice = new VeniceNode({
  apiKey: 'your-api-key',
});

// Generate a chat completion
async function generateChatCompletion() {
  const response = await venice.chat.createCompletion({
    model: 'llama-3.3-70b',
    messages: [
      { role: 'system', content: 'You are a helpful assistant' },
      { role: 'user', content: 'Tell me about AI' }
    ]
  });
  
  console.log(response.choices[0].message.content);
}

generateChatCompletion();

Documentation Sections

API Reference

Detailed documentation for all API resources and endpoints.

• Chat API
• Image API
• Models API
• API Keys
• Characters
• VVV Token

CLI Reference

Documentation for the command-line interface.

• Installation and Configuration
• Available Commands
• Advanced Usage
• Examples

Advanced Topics

Advanced usage and concepts.

• Streaming
• Error Handling
• Rate Limiting
• Debug Logging
• Function Calling
• Vision Models
• PDF Processing

Examples

Code examples for common use cases.

• Basic Chat
• Streaming Chat
• Generate Image
• Character Interaction
• Function Calling
• PDF Processing
• Vision & Multimodal

Special Features

Character Interaction

Learn how to interact with AI characters - predefined personalities that can enhance your conversational experiences.

Code Examples

Explore practical code examples for using the Venice Dev Tools SDK.

Troubleshooting

Common issues and solutions to help you resolve problems with the SDK.

Command Line Interface

Try our CLI for quick access to Venice AI capabilities

npm install -g venice-dev-tools
venice chat "Tell me about AI"

View CLI Documentation Explore Code Examples

Features

• Chat Completions: Generate text responses with streaming support and web search
• Image Generation: Create images with various models and styles
• Image Upscaling: Enhance image resolution
• Models Management: List models, traits, and compatibility mappings
• Character Management: List and interact with pre-defined AI characters
• API Key Management: Create, list, delete, and check rate limits for API keys
• VVV Token Information: Get circulating supply, network utilization, and staking yield
• Web3 Integration: Generate API keys using Web3 wallets
• Command Line Interface: Interact with the API directly from your terminal
• Error Handling: Comprehensive error handling with specific error classes
• Rate Limiting: Automatic rate limit tracking and handling
• Debug Logging: Robust logging system with multiple log levels and runtime configuration
• PDF Processing: Process PDF documents in different modes (binary data, text extraction, or both) with support for external PDF-to-image conversion
• Vision/Multimodal: Send both text and images to vision-capable models

Privacy-First Approach

Venice AI offers unparalleled privacy advantages over other AI providers:

• No Data Storage: Your prompts, responses, and generated content are never saved on any Venice infrastructure
• Local Storage Only: Your conversation history lives in your local browser - clear your browser data, and those conversations are gone forever
• Decentralized Processing: Your requests are processed on decentralized GPUs without any identifying information
• Transient Processing: Once a prompt is processed, it is purged from the GPU - nothing persists
• SSL Encryption: All communication is secured using SSL encryption throughout the entire journey

As Venice states: “You don’t have to protect what you do not have.”