OneAI Core
Commercial AI Operating Layer
Turn model APIs into paid, governed and measurable AI products with gateway access, task intelligence, routing policy, cost controls, usage tracking and a commercial console.
OneAI Core
Commercial Control Plane
OneAI Core
The flagship product behind the OneAI Labs ecosystem.
OneAI Core is the commercial layer above every model, task, customer and request. It lets teams route models, package structured intelligence, enforce policy, track usage and operate AI as a SaaS product.
OpenAI-compatible gateway
Use familiar chat completion calls while OneAI handles provider routing and product visibility.
Task Intelligence API
Package repeatable business workflows as typed task contracts with structured JSON outputs.
API key and customer control
Issue keys, scope access, separate environments and connect usage back to customers.
Usage, cost and latency tracking
Record provider, model, tokens, estimated cost, latency, requestId and error state.
Routing policy and cost guards
Apply cheap, balanced, fast, auto, premium or explicit provider:model behavior with maxCostUsd.
Billing-ready operations
Run AI like a SaaS product with plans, limits, commercial visibility and operator workflows.
Core architecture
One request. Multiple layers of commercial intelligence.
OneAI Core wraps every AI request with authentication, task shaping, routing, policy, provider execution, usage tracking and billing-ready visibility.
Your products keep their own UX. OneAI Core provides the infrastructure behind keys, routing, task contracts, limits, requests, costs and operator control.
Authenticate key
Validate API key, customer plan and access scope.
Shape task
Convert product intent into a stable task contract.
Apply routing
Select cheap, balanced, fast, auto or explicit provider:model behavior.
Enforce policy
Apply cost guard, tier access and request rules.
Execute model
Route to the selected provider, model catalog or gateway path.
Track usage
Record request, customer, usage, cost, latency and billing context.
Task Intelligence API
/v1/generate
Structured business workflows such as business_strategy, content_engine, support_brain and market_research.
Model Gateway API
/v1/chat/completions
OpenAI-compatible access with provider:model routing, streaming-ready behavior and observability.
Model Catalog
/v1/models
Catalog discovery, pricing coverage, readiness checks and model strategy for operators.
Developers
Two APIs. One product-grade AI system.
Use the gateway for familiar model access. Use Task Intelligence when your product needs stable business outputs instead of loose prompt responses.
/v1/generate
/v1/chat/completions
structured task api
curl https://oneai-saas-api-production.up.railway.app/v1/generate \
-H "content-type: application/json" \
-H "x-api-key: YOUR_API_KEY" \
-H "Idempotency-Key: launch-plan-001" \
-d '{
"type": "business_strategy",
"input": {
"goal": "Launch a paid AI feature in 30 days",
"audience": "SaaS founders and product teams"
},
"options": {
"llm": {
"mode": "balanced",
"maxCostUsd": 0.03
}
}
}'model gateway
curl https://oneai-saas-api-production.up.railway.app/v1/chat/completions \
-H "content-type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "openai:gpt-5.2",
"messages": [
{
"role": "user",
"content": "Turn this idea into a commercial AI product plan."
}
],
"max_completion_tokens": 300
}'Trust layer
Built like API infrastructure, not a toy demo.
OneAI Core is designed around production concerns: key hygiene, scoped access, cost control, request observability, provider policy and operator review.
Hashed API keys
Keep secrets server-side, create separate keys per environment and revoke leaked keys quickly.
Usage and cost visibility
Track provider, model, tokens, estimated cost, latency and requestId for support and operations.
Provider policy
Use routing modes, allowlists, fallbacks and maxCostUsd to keep production calls controlled.
Execution boundary
OneAI returns plans, structured decisions and coordination outputs while execution stays in product workflows.
Operational recommendation
Use idempotency keys, monitor usage daily and health-check new providers before exposing them to customers.
Build on the OneAI Labs commercial AI stack.
Start with OneAI Core for infrastructure, add Agent OS for workflows, and ship products that customers can use, trust and pay for.