Together AI vs Fikra API: Integration Comparison
Introduction
Together AI is an external AI infrastructure provider that offers access to various open-source and proprietary models. Fikra API is a separate service that provides access to its own set of models (such as fikra-flash and fikra-pro-20b) through an OpenAI-compatible interface. This page compares the integration approaches of these two distinct services to help developers understand how to connect to each.
Comparison
| Feature | Together AI | Fikra API |
|---|---|---|
| Integration Style | External provider with its own SDK and API structure. | OpenAI-style chat-completions endpoint. |
| Authentication | Uses Together AI API keys. | Uses Authorization: Bearer fk_live_... headers. |
| Endpoint Compatibility | Together AI specific endpoints. | Compatible with OpenAI SDKs via https://api.fikraapi.co.ke/v1. |
| Payment/Access | Managed through Together AI's billing system. | Managed via Fikra dashboard with Paystack integration (M-Pesa and cards accepted). |
| Model Naming | Uses Together AI's model identifiers. | Uses Fikra aliases like fikra-flash, fikra-fast-8b, fikra-pro-20b. |
Python Example: Using Fikra API with OpenAI SDK
Because Fikra API supports OpenAI-style requests, you can use the standard OpenAI Python SDK by pointing it to the Fikra base URL. Note that you must use a valid Fikra model alias.
import os
from openai import OpenAI
# Initialize the client with Fikra's base URL and API key
client = OpenAI(
base_url="https://api.fikraapi.co.ke/v1",
api_key=os.getenv("FIKRA_API_KEY")
)
# Make a request using a Fikra model alias
response = client.chat.completions.create(
model="fikra-flash",
messages=[
{"role": "user", "content": "Hello, Fikra!"}
]
)
print(response.choices[0].message.content)
When Fikra API is a Fit
Fikra API is a suitable choice for developers who:
- Prefer an OpenAI-compatible interface for easier integration with existing tools.
- Are based in or have access to payment methods supported by Paystack (such as M-Pesa).
- Want to use Fikra's specific model lineup, such as
fikra-flashfor speed orfikra-pro-20bfor higher capability. - Need a straightforward API key authentication flow without complex enterprise onboarding.
Verify Before Production
Before deploying Fikra API in a production environment, please:
- Check the Model Registry: Confirm that the specific model alias you intend to use is currently available and enabled for your account.
- Review Rate Limits: Be aware that unverified accounts are limited to 30 RPM, while trusted accounts are limited to 100 RPM. Ensure your application's traffic patterns fit within these limits.
- Test Streaming: If you plan to use
stream: true, verify that your client correctly handles Server-Sent Events (SSE). - Confirm Billing: Ensure your Fikra dashboard wallet has sufficient credits and that your payment method is correctly linked via Paystack.
Note: Fikra API does not host or resell models from Together AI or other external providers. Always verify model availability and specific terms directly with the provider.
James Miano
CTO & ML Engineer at Lacesse Ventures. James specializes in low-overhead LLM quantization processes, custom ternary weights architectures, and localized server optimization.
Stop paying for overpriced round-trip latency
Why route queries over Western servers when you can use low-overhead hardware located in Nairobi? Save 87% on your monthly inference spend. No minimum credit limits. M-Pesa ready.