AI

Availability: Preview

Prepaid, OpenAI-compatible inference by the token

Point the OpenAI SDK at ai.alawadi.cloud/v1, add an sk-alwd- key, and call qwen2.5-7b-instruct billed from your prepaid balance.

What you get

OpenAI SDK drop-in

Set base_url to https://ai.alawadi.cloud/v1 and your key; /v1/chat/completions and /v1/models work with no code rewrite.

Server-side token metering

Token counts come from the model, never from your client, so what you pay matches what you generated.

Portal-minted API keys

Create sk-alwd- keys in the portal; they hash at rest, show last-used, and revoke the instant you rotate them.

Qwen2.5-7B-Instruct, 32k context

One Apache-2.0 model with a 32,768-token context window and a strong Arabic baseline.

Runs on our own GPU in-region

Requests run on our own GPU in the UAE region, not a re-sold third-party API.

Prepaid from your balance

Draw from the same balance you top up with vouchers; no card and no monthly commit.

How it works

  1. 01

    Mint a key

    Sign in with Google, open the portal, and create an sk-alwd- API key.

  2. 02

    Point your SDK

    Set base_url to https://ai.alawadi.cloud/v1 and pass the key in any OpenAI client.

  3. 03

    Call and pay per token

    Send /v1/chat/completions requests; each call debits input and output tokens from your balance.

Pricing

Each service uses its published meter against prepaid credit: provisioned time, storage, tokens, or another listed unit. Estimate any setup in the calculator.

Open the pricing calculator
Input tokensCounted server-side, debited from your prepaid balance.
$0.10 / 1M tokens
Output tokensSame balance, billed per token, no card required.
$0.30 / 1M tokens
Welcome creditCovers your first calls; vouchers priced in SYP are the live top-up route today.
$5 free at signup

Questions

Which model can I call?

One model is live: Qwen2.5-7B-Instruct, served as qwen2.5-7b-instruct, Apache-2.0, with a 32,768-token context window.

Do you support embeddings or /completions?

No. Only /v1/chat/completions and /v1/models are live; /completions and embeddings return 404.

How am I billed?

Per token from your prepaid balance: $0.10 per 1M input tokens and $0.30 per 1M output tokens, counted server-side.

Do I need a credit card?

No. Claim the $5 welcome credit first; vouchers priced in SYP are the live top-up route today. Card and crypto top-ups are not live.

Related products

Deploy this in your first project.

Claim your $5 trial credit and go live today.

Start free