Getting started

DeepToken is a multi-model inference gateway. Point any OpenAI-compatible client at /v1 and pick a model from the catalog — we handle routing, fallback, and billing.

1. Create an account

Sign up at /signup. Email + password is fine; Google OAuth works too. New accounts start on the Free tier with a small monthly credit grant so you can ship your first call without entering payment info.

2. Mint an API key

In the dashboard, open API Keys → Create key. Copy the key once — we never show it again. Treat it like a password.

export DEEPTOKEN_API_KEY="dtk_..."

3. Make your first call

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 in one sentence."}]
  }'

That's it. The response shape matches OpenAI exactly, so existing SDKs work with only a base_url change.

Next steps

• Authentication — API key scopes, IP allowlists, per-key rate limits
• Models — the routable model catalog
• Errors — error categories and retry hints
• Pricing — credit packs, subscriptions, and per-model multipliers