LangDB

<Note type="warning"> This community provider is not yet compatible with AI SDK 5. Please wait for the provider to be updated or consider using an [AI SDK 5 compatible provider](/providers/ai-sdk-providers). </Note>

LangDB is a high-performance enterprise AI gateway built in Rust, designed to govern, secure, and optimize AI traffic.

LangDB provides OpenAI-compatible APIs, enabling developers to connect with multiple LLMs by changing just two lines of code. With LangDB, you can:

Provide access to all major LLMs
Enable plug-and-play functionality using any framework like Langchain, Vercel AI SDK, CrewAI, etc., for easy adoption.
Simplify implementation of tracing and cost optimization features, ensuring streamlined operations.
Dynamically route requests to the most suitable LLM based on predefined parameters.

Setup

The LangDB provider is available via the @langdb/vercel-provider module. You can install it with:

Provider Instance

To create a LangDB provider instance, use the createLangDB function:

tsx

import { createLangDB } from '@langdb/vercel-provider';

const langdb = createLangDB({
  apiKey: process.env.LANGDB_API_KEY, // Required
  projectId: 'your-project-id', // Required
  threadId: uuidv4(), // Optional
  runId: uuidv4(), // Optional
  label: 'code-agent', // Optional
  headers: { 'Custom-Header': 'value' }, // Optional
});

You can find your LangDB API key in the LangDB dashboard.

Examples

You can use LangDB with the generateText or streamText function:

`generateText`

tsx

import { createLangDB } from '@langdb/vercel-provider';
import { generateText } from 'ai';

const langdb = createLangDB({
  apiKey: process.env.LANGDB_API_KEY,
  projectId: 'your-project-id',
});

export async function generateTextExample() {
  const { text } = await generateText({
    model: langdb('openai/gpt-4o-mini'),
    prompt: 'Write a Python function that sorts a list:',
  });

  console.log(text);
}

generateImage

tsx

import { createLangDB } from '@langdb/vercel-provider';
import { generateImage } from 'ai';
import fs from 'fs';
import path from 'path';

const langdb = createLangDB({
  apiKey: process.env.LANGDB_API_KEY,
  projectId: 'your-project-id',
});

export async function generateImageExample() {
  const { images } = await generateImage({
    model: langdb.image('openai/dall-e-3'),
    prompt: 'A delighted resplendent quetzal mid-flight amidst raindrops',
  });

  const imagePath = path.join(__dirname, 'generated-image.png');
  fs.writeFileSync(imagePath, images[0].uint8Array);
  console.log(`Image saved to: ${imagePath}`);
}

embed

tsx

import { createLangDB } from '@langdb/vercel-provider';
import { embed } from 'ai';

const langdb = createLangDB({
  apiKey: process.env.LANGDB_API_KEY,
  projectId: 'your-project-id',
});

export async function generateEmbeddings() {
  const { embedding } = await embed({
    model: langdb.embeddingModel('text-embedding-3-small'),
    value: 'sunny day at the beach',
  });

  console.log('Embedding:', embedding);
}

Supported Models

LangDB supports over 250+ models, enabling seamless interaction with a wide range of AI capabilities.

Checkout the model list for more information.

For more information, visit the LangDB documentation.