Fikra API & Developer Documentation

API Overview & Architecture

The Fikra API is a standard RESTful interface designed to be an immediate drop-in replacement for OpenAI or Anthropic APIs. If you know how to build with global AI APIs, you already know how to build with Lacesse.

• Authentication: Secure your requests using standard Bearer Token authentication. Generate your unique API keys instantly in the Lacesse Developer Dashboard.
• Disruptive Pricing: Built for African scale. At just $0.002 per 1K inference tokens, Fikra provides enterprise-grade reasoning at a fraction of global market costs.
• Rate Limits & Scaling: Free tier users enjoy up to 60 Requests Per Minute (RPM). Paid tiers autoscale dynamically to support thousands of concurrent connections with 99.99% uptime.
• Zero Data Retention: By default, data passed through the API is not stored or used to train future Lacesse models, ensuring strict compliance with local data privacy laws.

Core Endpoints & Code Examples

Interact with the /v1/chat/completions endpoint to generate text, parse data, or simulate conversational memory. The API accepts standard JSON payloads.

cURL Example

curl -X POST "https://api.lacesse.co.ke/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_LACESSE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "fikra-7b-instruct",
    "messages": [
      {"role": "system", "content": "You are a helpful business assistant."},
      {"role": "user", "content": "How do I optimize my M-Pesa till flow?"}
    ],
    "temperature": 0.5,
    "max_tokens": 512
  }'
    

Python SDK Example

import os
from lacesse import FikraClient

client = FikraClient(api_key=os.environ.get("LACESSE_API_KEY"))

response = client.chat.completions.create(
    model="fikra-ternary-v1",
    messages=[{"role": "user", "content": "Jambo, how can I automate my inventory?"}]
)

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

Developer Integration Tutorials

Move from sandbox to production quickly with our official guides and open-source boilerplates.

• Integrating with React & Node.js: Learn how to set up Server-Sent Events (SSE) for streaming text responses directly to a web frontend.
• WhatsApp Business Chatbot Integration: A complete guide to connecting the Fikra API to Meta's WhatsApp API using webhooks to handle customer inquiries 24/7.
• Python SDK & LangChain Guide: How to use Fikra as your core LLM in LangChain for Retrieval-Augmented Generation (RAG) over PDF documents.
• JavaScript / TypeScript SDK Docs: Full type definitions and examples for Edge functions (Vercel/Cloudflare).

Developer FAQ & Troubleshooting

Common questions regarding API keys, limits, and integration workflows.