Frequently asked questions
What developers need to know before calling Yamify inference.
What is Yamify?
Yamify is building an OpenAI-compatible inference platform for African regions. Developers will call Qwen, Mistral, and Gemma from their own applications without operating GPU infrastructure. Production endpoints will identify their verified serving region.
What do I receive after activation?
You receive an API base URL, Bearer API key, model ID, serving region, usage meter, and copyable cURL and Python examples. There is no separate model dashboard login.
Do I need OpenClaw or n8n?
No. Your application can call the inference API directly. OpenClaw is an optional agent interface and n8n is an optional workflow consumer of the same API.
Is my data private?
Every endpoint will show its actual serving region. Do not treat a beta endpoint as an African residency guarantee until it displays verified African region evidence.
How fast can I get started?
The activation goal is one successful API completion. Approved builders receive setup instructions and can use an OpenAI-compatible SDK without rewriting their application.
What support do you offer?
Private-beta users receive founder-led onboarding and workload review. Dedicated capacity and production support are scoped after the workload is validated.
How do Sovereign AI Credits work?
Approved builders start with 100 Sovereign AI Credits. The dashboard shows remaining credits and usage. When credits reach zero, inference pauses rather than silently creating a bill.
Who qualifies for the private beta?
Teams with a concrete model, traffic, latency, and region requirement receive priority. Yamify reviews requests manually so scarce beta capacity goes to workloads that can produce useful product feedback.