Create account
Sign up, verify your email, and choose whether you are working solo or inside an organization.
Create an account, issue an API key, and send your first OpenAI-compatible request through the DeepToken gateway with routing, metering, and team controls enabled from day one.
Gateway base URL
https://api.deeptoken.app/v1Sign up, verify your email, and choose whether you are working solo or inside an organization.
Create a scoped key, set optional model and IP limits, then store the secret in your secret manager.
Point any OpenAI-compatible SDK at the /v1 gateway and pass a supported model id.
Review tokens, latency, fallback hops, costs, and request status from the dashboard.
Use budgets, organizations, scoped keys, and billing exports before production traffic grows.
DeepToken follows the OpenAI Chat Completions schema, so most existing SDK code only needs a new base URL and API key. The model id stays explicit while routing and fallback are handled by the gateway.
curl https://api.deeptoken.app/v1/chat/completions \
-H "Authorization: Bearer $DEEPTOKEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4o-mini",
"messages": [
{ "role": "user", "content": "Say hello from DeepToken" }
]
}'Use catalog model ids in your client code. DeepToken resolves the upstream provider, applies health-aware fallback, records token usage, and keeps your application isolated from provider-specific changes.
Production tip
Start with a primary provider plus at least one fallback for latency-sensitive traffic. Keep strict budgets on new keys until your usage pattern is stable.
Every request is metered into quota with model, key, route, and organization context. This keeps finance exports and engineering debugging aligned.