AI gatewayTask intelligenceCommercial control plane

The full-stack operating layer for model gateways, Task Intelligence, and Agent OS.

OneAI helps SaaS teams turn raw model access into paid, governed, measurable AI features with model routing, structured outputs, Agent OS handoff infrastructure, API keys, usage visibility, cost controls, and billing-ready operations.

OneAI is the intelligent coordination brain for model routing, Task Intelligence, and Agent OS handoff. OneClaw, bots, external agents, or human operators handle execution.

From model access to commercial AI infrastructure that can sell, measure, govern, audit, and customize intelligence.

oneai.product-preview
AI Operations Preview
Gateway, tasks, routing, policy, and commercial visibility in one operating layer.
PreviewSample flow
Gateway
ready
Chat completions
Messages
ready
Anthropic-style API
Tasks
ready
Structured outputs
Agent OS
preview
Handoff preview
Chat gateway ready
Messages ready
Task API ready
Agent OS preview
Product preview
Example interface only. Real requests, customers, costs, and usage live inside the authenticated dashboard.
Open Console
4
API surfaces
Generate + Chat + Messages + Agent OS
480+
model catalog target
Provider choice without product rewrites
4
routing modes
Cheap, balanced, premium, explicit
1
control plane
Keys, usage, customers, costs, billing
System Map

One request. Multiple layers of commercial intelligence.

Every AI request can move through authentication, task shaping, routing, cost policy, provider execution, usage tracking, and billing-ready visibility.

OneAI unifies model access, Task Intelligence, Agent OS handoff, and commercial controls into one operable intelligence brain.

Request lifecycle
System preview · not live production data
preview
01
Authenticate key
Validate API key, customer plan, and access scope.
02
Choose surface
Use Generate, Chat, Messages, or Agent OS preview based on the product need.
03
Shape intelligence
Convert product intent into a stable task, message, or handoff contract.
04
Apply routing
Select cheap, balanced, premium, or explicit model behavior.
05
Enforce policy
Apply cost guard, tier access, and request rules.
06
Call model or handoff
Route to a model provider or prepare a non-executing Agent OS handoff object.
07
Track usage
Record request, customer, usage, cost, and billing context.
Input
node
Your Product
SaaS app · backend · agent · internal tool
Models
node
Model Providers
OpenAI · Anthropic · Google · xAI · DeepSeek · OpenRouter
Workflow
node
Task Intelligence
business_strategy · content_engine · support_brain
Handoff
node
Agent OS Preview
agent plans · context packets · external execution boundary
Governance
node
Policy & Cost Guard
routing mode · maxCostUsd · plan gates
Commercial
node
Commercial Console
keys · usage · customers · billing · cost
Platform architecture

One commercial layer above every model, task, customer, and request.

OneAI combines model access, structured workflow intelligence, operational visibility, and commercial controls into one product layer.

01
Access

Model Gateway APIs

OpenAI-compatible Chat and Messages access with provider:model routing and product-level observability.

/v1/chat/completions
/v1/messages
streaming
model catalog
02
Intelligence

Generate Task API

Structured business workflows packaged as stable API contracts instead of loose prompts.

/v1/generate
typed task contracts
structured JSON
workflow outputs
03
Governance

Cost & Policy Layer

Control how AI is used with routing modes, cost guards, tier access, and request visibility.

maxCostUsd
routing modes
plan gates
request tracing
04
Agent OS

Preview & Handoff

Generate agent plans, context packets, and handoff contracts while keeping execution outside OneAI.

/v1/capabilities
/v1/agent-plans
/v1/handoff/preview
/v1/context/preview
05
Revenue

Commercial Console

Operate AI like a SaaS product with API keys, customers, usage, billing readiness, and analytics.

API keys
usage analytics
customers
billing readiness
Developer experience

Four API surfaces. One product-grade AI system.

Use Chat or Messages for familiar model access, Generate for structured Task Intelligence, and Agent OS endpoints for planning, context, and handoff previews.

oneai.api
structured task api
curl https://oneai-saas-api-production.up.railway.app/v1/generate \
  -H "content-type: application/json" \
  -H "x-api-key: $ONEAI_API_KEY" \
  -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 $ONEAI_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
  }'
Data boundary

OneAI returns intelligence and handoff contracts. Execution stays with OneClaw, bots, your system, or humans.

Cost control

Every request can carry model, token, latency, requestId, cost estimate, and routing trace metadata.

Commercial control

API keys, usage, customers, billing readiness, team roles, audit events, and Agent OS ledger live in one console.

Task Intelligence

Productize intelligence as reusable business contracts.

Task Intelligence gives your product stable inputs, structured outputs, plan-aware access, and measurable model cost.

Task
Use case
Tier
business_strategy
Executive-ready strategy
Free
content_engine
Product narrative and launch content
Free
campaign_mission
Growth campaign planning
Pro
support_brain
Customer support intelligence
Pro
market_research
Market and competitor brief
Pro
custom_task_designer
Customer-specific AI workflow
Team
Model strategy

Route by business need, not provider lock-in.

Cheap when possible. Premium when necessary. Explicit when customers require it. Observable every time.

OpenAI
frontier reasoning
gpt-5.2
Anthropic
judgment and long context
claude
Google
multimodal intelligence
gemini
xAI
realtime awareness
grok
DeepSeek
efficient structured logic
reasoner
OpenRouter
multi-provider reach
catalog routing
Why OneAI

The missing bridge between model APIs and monetized AI products.

Calling a model is easy. Turning that call into something customers can buy, trust, measure, and use repeatedly is the hard part.

Capability
Raw model API
OneAI
Integration
Wire every provider manually
One OpenAI-compatible gateway
Output
Prompt-shaped text
Task-shaped structured contracts
Cost control
Manual monitoring
Routing modes and cost guards
Operations
Scattered logs and usage
Keys, requests, customers, usage, cost
Commercialization
Build billing logic yourself
Plan-aware and billing-ready infrastructure
Trust layer

Built to be more than a pass-through model relay.

Customers do not only worry about model access. They worry about hidden routing, runaway cost, unclear data boundaries, and whether AI outputs are usable inside a real product.

Positioning

Generic gateways sell access to models. OneAI sells the operating layer that turns models into governed, measurable, task-specific business intelligence.

No black-box routing

Responses expose provider, model, tokens, requestId, latency, fallback state, and estimated cost so teams know exactly what happened.

Task contracts over loose prompts

Structured workflows package repeatable business logic behind typed inputs, stable outputs, tiers, and validation.

Cost-first operations

Routing modes, maxCostUsd, plan limits, usage dashboards, and model pricing help teams prevent AI spend drift.

Execution stays separate

OneAI coordinates intelligence. OneClaw, bots, or your own systems execute actions, keeping the API boundary clear.

Generic model gateway
OneAI intelligence layer
Primarily forwards requests to models
Turns business intent into reusable Task Intelligence
Competes on number of models
Competes on workflow quality, cost control, and commercial readiness
Often optimized for raw chat/completion calls
Supports model gateway calls plus structured business tasks
Trust depends on provider access
Trust comes from transparent routing, usage, cost, task contracts, and execution boundaries
Launch paths

Built for builders who need AI to become a product line.

OneAI is useful whether you are shipping your first AI feature or standardizing model access across a growing platform.

For SaaS founders

Launch AI features with keys, usage, model routing, and cost tracking without building the operating layer from scratch.

For AI product teams

Standardize model access, package repeatable workflows, enforce policies, and monitor usage across your product.

For agencies and builders

Turn strategy, content, research, support, and campaign workflows into reusable client-facing AI products.

Commercial control

Operate AI like a serious SaaS business.

OneAI gives you the commercial layer around AI: login, keys, requests, usage, customers, costs, routing policies, plan gates, and billing readiness.

Commercial Control Plane
Real operator data from OneAI customers, keys, usage, and events
OperatorLoading
Customers
-
Active keys
-
Requests
-
Model cost
-
Customer
Keys
Cost
Usage
Loading operator data...
Plan policy
Freecheap + balanced
Profast + auto
Teampremium + debug
Recent events
Loading events...
FAQ

Clear enough for builders. Serious enough for buyers.

Answer the most important product, technical, and commercial questions before users enter the console or docs.

Is OneAI a model provider?

No. OneAI sits above model providers as a commercial infrastructure layer. It helps your product route, govern, package, observe, and monetize AI usage.

Can I keep using OpenAI-compatible requests?

Yes. The gateway supports familiar /v1/chat/completions behavior, plus /v1/messages for Messages-style integrations, while adding provider routing, cost visibility, and operational controls.

What is Task Intelligence?

Task Intelligence turns repeatable business workflows into stable API contracts with defined inputs, structured outputs, access tiers, and cost visibility.

How is OneAI different from B.AI or a generic model gateway?

Those platforms are strongest at model access and payment rails. OneAI includes a gateway, but its higher layer is Task Intelligence: business-ready outputs, cost controls, request visibility, and commercial SaaS operations.

Are the console numbers on this homepage real?

No. The homepage console and system map are product previews. Real usage, cost, request, and customer data should come from the authenticated dashboard.

Start building

Build the AI infrastructure your customers can actually pay for.

Start with API access. Add routing, tasks, usage, cost control, customer visibility, and billing-ready operations as your AI product grows.