Send your first request

Steps#

1. Sign in with your wallet

   Open the console and connect an EVM wallet. We recommend Rabby; any browser wallet that supports EIP-6963 works. You sign a one-time message to prove you own the address. Signing is free and sends no transaction. Your address is your account.

2. Add balance

   Every request is paid from your balance, which is kept in US dollars, so it has to cover the request before it runs. On the Wallet page, deposit ETH or USDG on Robinhood Chain, or ETH or USDC on Base from the wallet you signed in with. It’s credited once the network confirms it: ETH at the US dollar rate at that moment, USDG and USDC 1:1, minus a 3% deposit fee. Gas is paid in ETH on the network you deposit from. Only deposits sent from your signed-in wallet are credited.

3. Create an API key

   Go to API keys → Create key with the inference scope. Keys are free. Pick the models the key may use: aliases such as inferiq/balanced, exact model ids, or all models. For this guide, allow inferiq/balanced. Copy the key; it starts with iq_, is shown once, and Inferiq stores only a hash. Optionally give the key a daily spending limit. Keep it out of browser code:

   .env

   Copy

   # .env: server side or your shell, never in browser code
   INFERIQ_BASE_URL=https://api.inferiq.run/v1
   INFERIQ_API_KEY=iq_••••••••

4. Send a request

   Every request names a model; there is no default. Use the alias inferiq/balanced and the gateway sends it to Qwen3.8-27B with medium reasoning, retrying once if it fails. If the model has been idle, the first request can take a few minutes while it wakes up. Any OpenAI-compatible client works, including the official SDKs.

   cURLTypeScriptPython

   Copy

   1234567891011curl $INFERIQ_BASE_URL/chat/completions \
     -H "Authorization: Bearer $INFERIQ_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
       "model": "inferiq/balanced",
       "messages": [
         {"role": "system", "content": "You are a concise coding assistant."},
         {"role": "user", "content": "Why does Array.prototype.map return undefined items?"}
       ],
       "max_tokens": 400
     }'

5. Read the response

   The body is a standard chat completion. Its model field and the x-inferiq-model header name the model that actually answered. The headers also carry the request ID and the receipt ID.

   Response headers

   Copy

   HTTP/1.1 200 OK
   content-type: application/json
   x-inferiq-request-id: req_QXt5TzjNYPURUd0
   x-inferiq-receipt-id: rcpt_QXt5TzjNYPURUd0
   x-inferiq-model: qwen/qwen3.8-27b

   200 OK · illustrative values

   Copy

   1234567891011121314151617181920212223{
     "id": "chatcmpl-msg_01H8x…",
     "object": "chat.completion",
     "created": 1792160000,
     "model": "qwen/qwen3.8-27b",
     "choices": [
       {
         "index": 0,
         "message": {
           "role": "assistant",
           "content": "map always returns an array of the same length. If your callback has no return…",
           "refusal": null
         },
         "finish_reason": "stop"
       }
     ],
     "usage": {
       "prompt_tokens": 52,
       "completion_tokens": 118,
       "total_tokens": 170,
       "prompt_tokens_details": { "cached_tokens": 0 }
     }
   }

6. Check the receipt

   Look the receipt up by its ID to see the tokens, the cost and your balance before and after. The same receipts are listed under Receipts in the console.

   terminal

   Copy

   curl $INFERIQ_BASE_URL/receipts/rcpt_QXt5TzjNYPURUd0 \
     -H "Authorization: Bearer $INFERIQ_API_KEY"

Choosing a model#

Deeper reasoning writes more tokens, so inferiq/best costs more per request than inferiq/fast. GET /v1/models lists only the models your key may use, with prices per 1M tokens. Read Models & fallback for the reasoning levels and when a request is retried.

Where to go next#

Point your coding tool at Inferiq (Connect your tool), learn how charges and limits work (Balance & receipts), and handle the error codes in your client (Errors).