One API key
Integrate once instead of managing credentials across six providers.
Smophy Mode API · Early access
One API. Six AI providers. The right model, chosen for you.
Send one request. SmophyAI automatically routes it to the best-suited model and returns the answer with the routing reasoning included.
ONE ENDPOINTONE API KEYAUTOMATIC ROUTINGROUTING METADATA
Routing across six leading AI providers
OpenAI
Anthropic
Perplexity
xAI
DeepSeek
One integration
Stop maintaining six provider integrations.
Six direct integrations mean six API keys, pricing systems, rate limits, SDKs, failure modes, and a permanent model-selection problem. Smophy Mode replaces them with one routing layer.
Intent-based routing
Match every request to the provider and model best suited to its task.
Provider health
Factor current availability and reliability into every routing decision.
Cost efficiency
Balance response quality and cost instead of defaulting to the most expensive model.
Request lifecycle
One request in. Routed answer out.
Send one prompt
Make one request through the Smophy Mode API integration.
Automatic routing
The engine classifies the task and selects the best-suited model across six providers.
Receive answer + reasoning
Get the model response together with which model was selected and why.
routing-result.json
{
"response": "…",
"routing": {
"provider": "best_suited_provider",
"model": "best_suited_model",
"reasoning": "task-specific routing explanation"
}
}Illustrative response shape only. Endpoint, authentication, exact request and response schema, and rate limits are provided to approved partners.
Fully automatic
There is no parameter to force a specific model. Every request uses the same intent-based routing as Smophy Mode in the app.
Pay as you go
Cost depends on the selected model, plus infrastructure and platform costs. Usage is billed from the first request; there is no free API tier.
Up to five keys
Approved accounts can create up to five keys from the SmophyAI side panel, separating development, production, and projects.
Routing engine
The same logic, visible in the app.
Before integrating, explore real routing examples with the detected task, confidence, capability score, provider health, cost efficiency, and alternative models.
See Smophy Mode in practiceIntent classification with confidence score
Per-task capability matching
Historical model track record
Real-time provider health
Quality and cost-efficiency balance
Routing explanation returned with the answer
FAQ
Smophy Mode API questions
What models does the Smophy Mode API route between?
The API routes across the top models from six AI providers: OpenAI (GPT), Anthropic (Claude), Google (Gemini), Perplexity, xAI (Grok), and DeepSeek.
How is API pricing calculated?
Pricing is based on the model your request is routed to, plus infrastructure and platform costs. There is no flat, single rate; cost depends on which model handles each request.
Can I force a specific model instead of using automatic routing?
No. Routing is fully automatic, with no request parameter to force a specific model. Every request goes through the same intent-based routing used in the app.
Is there a free tier or trial for the API?
No. The API has no free tier; usage is pay-as-you-go starting from your first request.
How do I get an API key?
The API is currently available to select partner companies. Request an invite through the form on this page, and once approved, generate your keys from the side panel in your SmophyAI account, up to 5 per account.
Build on the routing layer.
The API is open to select partner companies. Tell us what you’re building and we’ll reach out when a spot opens.
Request early access