vantedge-ai-sdk

VantEdge AI SDK provider for AI SDK - A custom provider that implements OpenAI-compatible endpoints for AI SDK.

Installation

npm install vantedge-ai-sdk

Usage

Basic Usage

import { vantedgeai } from 'vantedge-ai-sdk';
import { generateText } from 'ai';

const { text } = await generateText({
  model: vantedgeai('your-model-id'),
  prompt: 'Hello, how are you?',
});

Chat Model

import { vantedgeai } from 'vantedge-ai-sdk';
import { streamText } from 'ai';

const { textStream } = await streamText({
  model: vantedgeai.chatModel('your-chat-model-id'),
  messages: [{ role: 'user', content: 'Hello, how are you?' }],
});

Completion Model

import { vantedgeai } from 'vantedge-ai-sdk';
import { generateText } from 'ai';

const { text } = await generateText({
  model: vantedgeai.completionModel('your-completion-model-id'),
  prompt: 'Complete this sentence: The weather today is',
});

Embedding Model

import { vantedgeai } from 'vantedge-ai-sdk';
import { embed } from 'ai';

const { embedding } = await embed({
  model: vantedgeai.textEmbeddingModel('your-embedding-model-id'),
  value: 'This is a sample text to embed.',
});

Image Model

import { vantedgeai } from 'vantedge-ai-sdk';
import { generateImage } from 'ai';

const { image } = await generateImage({
  model: vantedgeai.imageModel('your-image-model-id'),
  prompt: 'A beautiful sunset over the ocean',
});

Configuration

Custom Provider Instance

import { createVantEdgeAI } from 'vantedge-ai-sdk';

const customProvider = createVantEdgeAI({
  apiKey: 'your-api-key',
  baseURL: 'https://your-api-endpoint.com/v1',
  headers: {
    'Custom-Header': 'custom-value',
  },
  queryParams: {
    version: '2024-01-01',
  },
});

// Use the custom provider
const { text } = await generateText({
  model: customProvider('your-model-id'),
  prompt: 'Hello world',
});

Environment Variables

You can set the API key using environment variables:

export VANTEDGE_API_KEY="your-api-key"

API Reference

createVantEdgeAI(options)

Creates a custom VantEdge AI provider instance.

Options

• apiKey (string, optional): Your VantEdge AI API key
• baseURL (string, optional): Base URL for API calls (default: environment-specific)
• headers (Record<string, string>, optional): Custom headers to include in requests
• queryParams (Record<string, string>, optional): Custom query parameters
• fetch (FetchFunction, optional): Custom fetch implementation

vantedgeai

Default provider instance with default configuration.

Model IDs

The package supports various model types:

• Chat Models: Use with vantedgeai() or vantedgeai.chatModel()
• Completion Models: Use with vantedgeai.completionModel()
• Embedding Models: Use with vantedgeai.textEmbeddingModel()
• Image Models: Use with vantedgeai.imageModel()

Features

• ✅ OpenAI-compatible API endpoints
• ✅ Support for chat, completion, embedding, and image models
• ✅ Customizable headers and query parameters
• ✅ Environment variable support for API keys
• ✅ TypeScript support with full type definitions
• ✅ Custom fetch implementation support
• ✅ Thread ID management for conversation context

Requirements

• Node.js 18+
• AI SDK 1.0+

License

Apache 2.0

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.