Introduction

SovereignEG is the MENA region's AI inference API. Send an HTTP request with a prompt, get tokens back through one OpenAI-compatible API, and pay in Egyptian pounds.

Access today, sovereignty next. Standard requests route through vetted model providers; contact us for Egypt-hosted deployments when data residency is required.

What you can do

  • Chat completions — text generation, reasoning, code, Arabic & multilingual conversation
  • Embeddings — vector representations for search, RAG, and classification
  • Pre-built agents — coming soon; not part of the generally available API yet
  • Custom agents — coming soon; build against the OpenAI-compatible API today

Key concepts

ConceptDescription
API KeyA secret token (sk-xxx) that authenticates your requests. Create one in the dashboard.
Model IDA string from GET /v1/models that selects which model processes your request.
TokensThe unit of text processing. Roughly 1 token ≈ 4 characters in English, ~2 characters in Arabic.
Base URLhttps://sovereigneg.com/v1 — OpenAI-compatible base URL for all API requests.

Quick example

from openai import OpenAI
 
client = OpenAI(
    api_key="sk-...",
    base_url="https://sovereigneg.com/v1"
)
 
response = client.chat.completions.create(
    model="...",  # live id from GET /v1/models
    messages=[
        {"role": "user", "content": "Hello from Cairo!"}
    ]
)
 
print(response.choices[0].message.content)

Yes — that's the standard OpenAI SDK. Change base_url, and you're running on SovereignEG. Zero code changes.

Why SovereignEG?

SovereignEGUS-hosted providers
API compatibilityOpenAI-compatible /v1 endpointsNative provider-specific APIs
BillingEGP credit balance and invoicesUSD with FX exposure
CatalogLive model list with EGP per-1M-token ratesProvider-specific pricing pages
RoutingTransparent standard routing today; Egypt-hosted option on requestUS/EU-hosted defaults
GovernanceOrganizations, projects, keys, limits, and audit trailsVaries by provider

Next steps

NextQuickstart