Anyscale Endpoints Alternative: Fikra API Integration Guide

Introduction

Anyscale Endpoints is a platform for deploying and serving machine learning models, often used for custom model hosting and inference. Fikra API is a separate service that provides access to specific Fikra model aliases through an OpenAI-compatible interface. This page compares the integration approaches for developers considering Fikra API as an alternative for OpenAI-style chat completions, without implying that Fikra hosts or resells Anyscale models.

Comparison

Python Example

Below is an example of using the OpenAI Python SDK to interact with Fikra API. Ensure you have the openai package installed and your API key stored in an environment variable.

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.fikraapi.co.ke/v1",
    api_key=os.getenv("FIKRA_API_KEY")
)

completion = client.chat.completions.create(
    model="fikra-flash",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ]
)

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

When Fikra API is a Fit

Fikra API is suitable for developers who: - Require an OpenAI-compatible endpoint for chat completions. - Prefer using specific Fikra model aliases (fikra-flash, fikra-fast-8b, fikra-pro-20b, fikra-pro-120b). - Need a token-credit based billing model with Paystack integration. - Are looking for a straightforward integration without custom model deployment.

Verify Before Production

Before deploying to production, please: 1. Check the Model Registry: Confirm the availability and specific capabilities of the desired Fikra model alias. 2. Review Rate Limits: Note that unverified accounts are limited to 30 RPM, while trusted accounts are limited to 100 RPM. 3. Test Authentication: Ensure your FIKRA_API_KEY is correctly configured and valid. 4. Validate Billing: Confirm your token-credit wallet balance and payment method setup.

For the most current information on models, pricing, and features, please refer to the official Fikra API documentation at https://fikraapi.co.ke.