U API

Uhuru API

Build AI-powered apps with Africa's intelligence. Drop-in SDK compatible.

Uhuru API

Build AI-powered applications with Uhuru's Chat Completions API. It implements the standard chat-completions wire format, so any SDK built for that format works by changing only the base URL and API key.

Quick Start

1

Generate an API key

Go to → API Keys and create a new key.

2

Make your first request

bash
curl -X POST https://ucloud.orionx.xyz/functions/v1/uhuru-public-api/v1/chat/completions \
  -H "Authorization: Bearer uhuru_sk_YOUR_KEY_HERE" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uhuru-4",
    "messages": [
      {"role": "user", "content": "Hello from Africa!"}
    ]
  }'
3

Or use a standard Python SDK

python
from openai import OpenAI

client = OpenAI(
    api_key="uhuru_sk_YOUR_KEY_HERE",
    base_url="https://ucloud.orionx.xyz/functions/v1/uhuru-public-api/v1",
)

response = client.chat.completions.create(
    model="uhuru-4",
    messages=[{"role": "user", "content": "Hello from Africa!"}],
)
print(response.choices[0].message.content)

Key Features

African Context Awareness

Regionally optimised responses with support for Setswana, Swahili, and more.

Drop-In Compatible

Implements the standard chat-completions format — existing Python & Node SDKs work unchanged.

Streaming Support

Real-time SSE streaming with data: [DONE] termination.

Secure by Default

SHA-256 hashed API keys, per-minute + monthly rate limiting, HTTPS only.

Base URL

https://ucloud.orionx.xyz/functions/v1/uhuru-public-api/v1

Available Models

ModelDescription
uhuru-4.5Deep-reasoning model — thinks before answering. Best for hard reasoning, analysis and long-form work. Up to 16k output tokens.
uhuru-4Fast general model for chat and high-volume work. Up to 8k output tokens. Default model.
uhuru-3.5Legacy compatibility ID — routes to the current fast general model.
uhuru-3.1Legacy compatibility ID — routes to the current fast general model.

Rate Limits

TierPer MinutePer Month
Free10 requests1,000 requests
Essential10 requests10,000 requests
Pro10 requests100,000 requests
Plus10 requests500,000 requests

Rate limit headers are included in every response: X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset.