Use Digixai with any OpenAI SDK.

Create a key in your dashboard. Send it as a bearer token on every request. Keys start with dgx- and are shown only once.

Authorization: Bearer dgx-xxxxxxxxxxxxxxxx

POST /chat/completions — identical request and response shape to OpenAI.

curl https://your-domain/api/public/v1/chat/completions \
  -H "Authorization: Bearer $DIGIXAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ant/claude-sonnet-5",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://your-domain/api/public/v1", apiKey: process.env.DIGIXAI_API_KEY });
const r = await client.chat.completions.create({
  model: "ant/claude-sonnet-5",
  messages: [{ role: "user", content: "Hello!" }],
});

Set "stream": true to receive server-sent events. Token usage is billed when the stream finishes.

curl https://your-domain/api/public/v1/models

Returns every enabled model with Digixai pricing per 1M tokens.

Digixai is prepaid. Each request is charged for the tokens actually used: input × input price + output × output price.

Before a request runs, max_tokens is lowered automatically so the cost can never exceed your balance or the key's spend limit.

• 401 — invalid or revoked key
• 402 — insufficient credits or key spend limit reached
• 403 — account suspended
• 404 — model not available
• 502 — upstream provider error