Integrating Fikra API with Vercel AI SDK

This guide demonstrates how to use the Vercel AI SDK with Fikra API by configuring it as a custom OpenAI-compatible endpoint. This allows you to leverage Fikra's models while utilizing the robust features of the Vercel AI SDK.

Prerequisites

1. Install the necessary packages: bash npm install ai @ai-sdk/openai

2. Ensure you have a valid Fikra API key.

Step 1: Configure the OpenAI Provider

To use Fikra API, you need to create a custom OpenAI provider instance that points to the Fikra base URL. The Fikra API is OpenAI-compatible, so we can use the createOpenAI function from @ai-sdk/openai.

import { createOpenAI } from '@ai-sdk/openai';

// Create a custom OpenAI provider for Fikra API
const fikra = createOpenAI({
  baseURL: 'https://api.fikraapi.co.ke/v1',
  apiKey: process.env.FIKRA_API_KEY,
});

Step 2: Use the Fikra Model

Now you can use the Fikra model with the Vercel AI SDK. Replace fikra-model-name with the specific model identifier provided by Fikra.

import { generateText } from 'ai';

const { data } = await generateText({
  model: fikra('fikra-model-name'),
  prompt: 'Hello, how are you?',
});

console.log(data);

Step 3: Streaming Responses

You can also stream responses from the Fikra API using the streamText function.

import { streamText } from 'ai';

const { stream } = streamText({
  model: fikra('fikra-model-name'),
  prompt: 'Tell me a story.',
});

// Consume the stream
for await (const chunk of stream) {
  process.stdout.write(chunk);
}

Environment Variables

Make sure to set the FIKRA_API_KEY environment variable in your .env file:

FIKRA_API_KEY=your_fikra_api_key_here

Conclusion

By following these steps, you can seamlessly integrate Fikra API with the Vercel AI SDK, enabling you to build powerful AI applications with minimal configuration.