FowyldAI Documentation
Everything you need to deploy, configure, and operate sovereign AI infrastructure.
What is FowyldAI?
FowyldAI is a sovereign intelligence engine that runs entirely on your infrastructure. No cloud calls, no data exfiltration, no vendor lock-in. Your data never leaves your environment — that's not a feature, it's the architecture.
5 min
Quickstart
Deploy FowyldAI and run your first query in under five minutes.
Reference
API Reference
Complete endpoint documentation with request/response examples.
Guide
Deployment
Docker, Kubernetes, and air-gapped deployment options.
Architecture
Data Sovereignty
How FowyldAI enforces zero-exfiltration at every layer.
Quick Install
Early Access
FowyldAI is currently in early access. Your private registry URL and API key are provided after signup. The commands below use that URL — replace
YOUR_REGISTRY with the value from your welcome email.
Docker (recommended)
docker pull YOUR_REGISTRY/fowyldai/engine:latest
docker run -d -p 8000:8000 --gpus all YOUR_REGISTRY/fowyldai/engine:latestPython package
pip install fowyldai
fowyldai serve --port 8000Verify
curl http://localhost:8000/health{
"status": "healthy",
"version": "1.5.0",
"models_loaded": 1,
"sovereign": true
}Core Concepts
| Concept | Description |
|---|---|
| Orchestration Engine | Multi-model orchestrator that routes queries to the optimal local model based on complexity, domain, and resource availability. |
| Sovereignty | All processing happens on your infrastructure. No telemetry, no cloud calls, no data leaves the boundary. |
| Deliberation Framework | Multiple models deliberate on complex queries, cross-validating responses for accuracy. |
| Air-Gap Mode | Full operation with zero network access — models, config, and updates delivered via sealed packages. |
| OpenAI Compatibility | Drop-in replacement for OpenAI API — use existing tools and libraries without code changes. |
API at a Glance
POST
/ask
Send a query, get a reasoned response
GET
/health
Engine health and model status
GET
/models
List loaded models and capabilities
POST
/v1/chat/completions
OpenAI-compatible chat endpoint
POST
/v1/embeddings
Generate text embeddings locally
GET
/sovereignty/status
Verify zero-exfiltration compliance
System Requirements
| Component | Minimum | Recommended |
|---|---|---|
| CPU | 8 vCPU | 16+ vCPU |
| RAM | 32 GB | 64 GB |
| Storage | 100 GB | 500 GB SSD |
| GPU | Optional (CPU fallback) | NVIDIA with 8+ GB VRAM |
| OS | Ubuntu 22.04+, RHEL 9+ | Ubuntu 24.04 LTS |
| Network | None required | Internal LAN only |
Editions
| Edition | Users | Key Features | Price |
|---|---|---|---|
| Starter | 1 | Crown Engine, unlimited local inference, basic workflows | $49/mo |
| Pro | Up to 5 | Multi-agent workflows, Institutional Memory, Admin Console | $149/mo |
| Business | Up to 20 | Governance, audit logging, Gateway & MCP, cloud augmentation | $399/mo |
| Enterprise | Unlimited | Custom integrations, compliance support, dedicated deployment | Contact Sales |
Annual billing saves 17%. See full pricing for details.