5-minute start

From signing up to your first model call — in five minutes.

1. Sign up

Open primerouter.ai → Sign Up. Available methods:

• email + password
• OAuth: GitHub, Discord, LinuxDO, etc.

New accounts come with a small starter quota — enough to make your first request.

2. Top up (skip if not needed)

Skip this step if the starter quota is enough. Otherwise:

• Crypto (recommended): Console → Top-up → pick chain & token, send to your dedicated address. Auto-credited on confirmation. See crypto user guide.
• Cards / Apple Pay: via Stripe.
• Redemption code: Console → Redemption → paste code.

3. Create an API key

Console → Tokens → New:

• Name: anything for your own reference
• Model restriction: leave empty for all models
• Expiry: leave empty for never

Save and copy the sk-... string — this is the only time the full key is shown. Lose it and you'll have to recreate.

4. First call

The simplest curl example:

curl https://primerouter.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [
      {"role": "user", "content": "Hello, please introduce yourself"}
    ]
  }'

Expected response:

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "deepseek-chat",
  "choices": [{
    "index": 0,
    "message": {"role": "assistant", "content": "..."},
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 28,
    "total_tokens": 40
  }
}

The usage field is byte-identical to upstream. Your bill is computed from this.

5. View usage and bills

Console → Logs: one row per request, click to expand:

• upstream provider & model
• token consumption
• actual quota deducted
• trace-id (matches the response header)

Now you've completed the smallest primerouter loop.

Next

• First call — streaming / multimodal / tool use
• Migrate from OpenAI SDK
• API docs
• Coding tools — one-click setup for Claude Code / Codex / Cursor and more