Integrating Fikra API with AnythingLLM

Prerequisites

• A Fikra API account with a valid API key (fk_live_...).
• Access to the Fikra dashboard to manage keys and token credits.
• AnythingLLM installed and running locally or in your environment.
• Basic familiarity with environment variables and JSON configuration.

Configuration Overview

AnythingLLM supports custom OpenAI-compatible endpoints. Since Fikra API exposes an OpenAI-style chat-completions interface, you can configure AnythingLLM to use Fikra as a custom LLM provider.

Step 1: Set Environment Variables

Export your Fikra API key as an environment variable. This ensures the key is not hardcoded in configuration files.

export FIKRA_API_KEY="fk_live_your_key_here"

Step 2: Configure AnythingLLM

In AnythingLLM, navigate to Settings > LLM Providers and select Custom OpenAI (or equivalent custom endpoint option).

Enter the following values:

• Base URL: https://api.fikraapi.co.ke/v1
• API Key: Your Fikra API key (fk_live_...)
• Model: fikra-flash

Note: Check the Fikra model registry in your dashboard for the latest available aliases. fikra-flash is used here as a default example.

Python Example

Below is a minimal Python example using the openai Python SDK, which is compatible with Fikra’s OpenAI-style interface. This can be used to verify connectivity before configuring AnythingLLM.

import os
from openai import OpenAI

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

response = client.chat.completions.create(
    model="fikra-flash",
    messages=[
        {"role": "user", "content": "Hello, Fikra!"}
    ]
)

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

Streaming Guidance

Fikra API supports streaming via Server-Sent Events (SSE) when stream: true is set. In AnythingLLM, enable streaming in the LLM provider settings if available. For custom integrations, use the stream parameter in your request.

Python Streaming Example

import os
from openai import OpenAI

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

stream = client.chat.completions.create(
    model="fikra-flash",
    messages=[
        {"role": "user", "content": "Explain quantum computing in simple terms."}
    ],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Common Errors

Production Checklist

• [ ] Store FIKRA_API_KEY in a secure environment variable or secrets manager.
• [ ] Use https://api.fikraapi.co.ke/v1 as the base URL.
• [ ] Select an appropriate Fikra model alias (e.g., fikra-flash) based on your use case.
• [ ] Implement retry logic with exponential backoff for 429 and 503 errors.
• [ ] Monitor token-credit usage in the Fikra dashboard to avoid 402 errors.
• [ ] Test streaming behavior if your application requires real-time responses.
• [ ] Verify rate limits (30 RPM unverified, 100 RPM trusted) and adjust concurrency accordingly.
• [ ] Log and alert on non-2xx responses for operational visibility.

Disclaimer: Fikra API is an OpenAI-compatible interface. Always refer to the official Fikra documentation and dashboard for the latest model availability, pricing, and rate limits.