Keep your SDK. Change the base URL.

1 · curl

curl https://oevuai.com/v1/chat/completions \
  -H "Authorization: Bearer $OEVAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Caelim-2.1-Pro",
    "messages": [
      {"role": "user", "content": "Hello, who are you?"}
    ],
    "stream": true
  }'

2 · Python (openai SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://oevuai.com/v1",
    api_key=os.environ["OEVAI_API_KEY"],
)

resp = client.chat.completions.create(
    model="Caelim-2.1-Pro",
    messages=[{"role": "user",
               "content": "Hello"}],
    stream=True,
)

3 · List models

curl https://oevuai.com/v1/models \
  -H "Authorization: Bearer $OEVAI_API_KEY"

4 · Check balance

curl https://oevuai.com/v1/user/balance \
  -H "Authorization: Bearer $OEVAI_API_KEY"

Authentication

Send your key as a bearer token. Keys are created and revocable in the platform console. Never ship a key in client-side code — use the embeddable widget for browser-side chat.

Rate limits

Requests are rate-limited per account; when you exceed the limit the API responds with a standard rate-limit error — back off and retry. Limits are shown in the platform console alongside usage.

Pricing

Free during the open beta. Per-million-token input/output pricing is published per model in the console so there are no surprises when billing starts.

Errors

Responses follow the OpenAI error shape: a JSON object with error.message and error.type. HTTP status codes are conventional (401 auth, 429 rate limit, 5xx server).