WORKSPACE

Your next integration.

Manage credentials, inspect usage and read the request reference.

Start with a key

Store keys on your server. Never embed a secret in browser code or a public repository.

API keys

Keys belong to your account. Secrets are shown once and expire after 90 days.

Playground

Send a text request using one of your account’s API keys.

Usage

Last 30 days. Rejected requests consume no tokens and incur no charges.

Credits

Payments are not enabled. There is no checkout or automatic billing.

API reference

Complete API guide: setup, payloads and error handling

Base URL: https://api.ramzai.org

Quick start

1. Verify your email and create a key in API keys.
2. Copy the secret when it is shown. Store it on your server as RAMZ_API_KEY; do not put it in browser code.
3. Send the bearer token with every request. Revoke exposed or unused keys.

Authentication

Send Authorization: Bearer YOUR_API_KEY. Invalid, expired or revoked keys return 401.

Chat request

POST /v1/chat/completions
Content-Type: application/json
Authorization: Bearer YOUR_API_KEY

{"model":"ramz0","messages":[{"role":"system","content":"Answer clearly and concisely."},{"role":"user","content":"Explain speech recognition."}]}

The model field is the requested model identifier. messages holds the conversation in order; each message has role and content.

curl https://api.ramzai.org/v1/chat/completions \
  -H "Authorization: Bearer $RAMZ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ramz0","messages":[{"role":"user","content":"Hello"}]}'

Credit check

Each account gets ten successful free chat requests, shared across every key and the playground. Each request allows up to 500 output tokens. Creating a new key does not reset this allowance. After it is exhausted, insufficient credit returns HTTP 402:

{"error":{"type":"insufficient_credit","code":"insufficient_credit","message":"Your account has insufficient credits."},"request_id":"REQUEST_ID"}

Rejected zero-credit requests have no inference charge. The request_id is also returned in x-request-id. Check your balance in Credits and request records in Usage. Do not retry 401 or 402 repeatedly without addressing the cause.

Read answers from choices[0].message.content and token counts from usage. The service limits output to 500 tokens; do not send max_tokens. At most two requests run concurrently per account. Invalid parameters return 400; rate limits return 429; generation failures return 502 and restore the free allowance. Streaming is not supported.

Model catalog

GET /v1/models returns the actual available models.

curl https://api.ramzai.org/v1/models \
  -H "Authorization: Bearer $RAMZ_API_KEY"

Use data[].id when selecting a model from the returned catalog. The catalog is not an inference response.

Loading catalog…

Settings

Profile

Security

Session

Login is separate from RAMZ AI administration. Signing out ends your browser session. Already issued API access tokens may remain valid for up to 15 minutes; API keys remain active until revoked or expired.