Skip to main content
Starrise AI is an enterprise AI API aggregation platform. Use one API key and one Base URL to call leading models and manage usage, billing, and call logs in one place.
Model lists, pricing, routes, supported client versions, and configuration paths may change with provider policies and platform updates. For production use, the current Console, your quotation, the SLA, data processing terms, and signed documents take precedence.

​
Product

Starrise is an enterprise AI API aggregation platform. Use one API key and one Base URL to call leading models — including Claude, GPT/OpenAI, Gemini, DeepSeek, Qwen, GLM, Kimi, Grok, and Doubao — and manage usage, billing, and call logs in one place.Core capabilities: unified access, intelligent routing and failover, transparent usage and cost, enterprise key and permission management, and 24/7 support from a UAE-based team.Chat, image generation, and video generation are all available through a single, consistent interface. See the Introduction and Quick Start.
Integrate once to access multiple leading models, and manage keys, permissions, usage, and billing from one console. You avoid separate contracts, integrations, and reconciliation for each provider, and you are not locked in to any of them.Compared with other aggregators, Starrise gives you 24/7 support from a UAE-based team, international and Chinese models through one API, business-ready payment (bank transfer, multiple currencies, formal invoices), and enterprise discounts based on your models and usage.

​
Getting started

Register at ai.alad.com, then go to Token Management in the Console to create your API key.Keep the key secure. Do not expose it in client-side code or public repositories.
Sign up at ai.alad.com and create an API key in the Console. Then replace the Base URL and API key in your existing OpenAI client with your Starrise values.The API is OpenAI-compatible, so no major code changes are needed.
  • Base URL: https://ai.alad.com/v1
  • Chat endpoint: POST https://ai.alad.com/v1/chat/completions
Copy the exact address from the Console if needed. Use model names exactly as listed in the Starrise marketplace, or requests will fail.See the Quick Start and API Reference for a complete first-call example.
Yes. Chat models fully conform to the OpenAI API specification. Change base_url to https://ai.alad.com/v1 — no other code changes are needed.
You can also list models available to the current key via GET /v1/models. See GET /v1/models.
Yes. Different models correspond to different channels. You must enable the corresponding channel in the Console before you can successfully call that model.

​
Models

Starrise supports models from multiple providers, including Anthropic, OpenAI, Google, ByteDance, DeepSeek, Alibaba, Zhipu, MiniMax, MoonShot, xAI, Kuaishou, and Black Forest Labs.Browse the full list in the Model API section, or call GET /v1/models with your API key. The marketplace in the Console is the source of truth for currently available model names.
Image and video models use dedicated endpoints rather than /v1/chat/completions:
  • Image generation: POST /v1/images/generations
  • Video generation: POST /v1/video/generations, then poll GET /v1/video/generations/{task_id} until the task succeeds
  • Provider-specific video APIs (for example Kling) use their own paths under /kling/v1/...
Video generation is asynchronous. Submit a request, then poll the task until the status becomes succeeded. See the API Reference for endpoint details and the Model API pages for each model’s parameters.

​
Billing

Billing is prepaid and usage-based. Top up your balance and each call is deducted at the model’s rate, with no monthly fee.
  • Chat models are billed per token (input + output)
  • Image and video generation models are billed per generation
  • Unused credits never expire
  • There are no platform fees or hidden markups
Listed prices match each provider’s official list price, so you can compare directly. Rates vary by model: check the marketplace or your quotation. Enterprise customers can contact us for discounted rates.One balance and one bill across providers, plus switching models without re-integrating, also cut your admin and switching costs.Go to the Top Up page to add credits. See Top Up & Payment.
API calls will return a 402 Payment Required error. Top up at the Console to resume usage.Set a balance alert to avoid interrupted API calls. We recommend a threshold of 20% of your average monthly spend.
Everything is in the Console. The dashboard shows your total spend, usage logs break down usage and trends by model, and the billing page records top-ups.To control cost, use multiple keys and groups, quotas, permissions, and balance alerts. We recommend a separate key for each environment and team, and exporting call details regularly to reconcile your bill.
Enterprise top-ups are processed manually by our operations team and are normally credited within 1 working day.You can pay by corporate bank transfer, with settlement in USD, HKD, AED, and other currencies. We send the receiving account and payment details with your invoice. Confirm them with our team before paying, and never use account details from unverified channels.Individual accounts can also add credits by credit or debit card via Stripe on the Top Up page.
Yes. We can provision trial credit for teams evaluating Starrise. To request it, tell us your use case, estimated volume, target models, expected concurrency, and preferred testing window.

​
Account and security

Go to Token Management in the Console to create, view, and disable keys, and assign groups, quotas, and permissions per project, team, or customer.Each key can have its own expiry, spending limit, and IP allowlist. Keys are masked in the Console and never appear in clear text in logs. Include the key as a Bearer token:
You can set an expiry, spending limit, and IP allowlist on each API key. Keys are masked in the Console and never appear in clear text in logs.Use sub-accounts to separate teams: a parent account sees its own and its sub-accounts’ data, and a sub-account sees only its own. Accounts support two-factor authentication, passkeys, and enterprise SSO.
We do not store request content. Prompts, responses, and uploaded files are held in memory only while the request is processed, then released. They are never written to disk or a database, and there is no option to log full request content.We retain only call metadata (model, token counts, latency, route, request ID), billing records, and an admin audit trail. Source IPs and error messages are not logged by default; if you enable error logging, they are redacted first.Training use, data residency, and deletion depend on the route, the provider’s policy, and your contract. Confirm the applicable setup and complete your compliance review before going to production.

​
Clients and tools

Use a tool such as CC Switch to route requests from Claude desktop, Codex, and similar clients to Starrise, or configure the third-party API directly in the client’s developer / custom-provider settings.Typical settings:
  • API Key: your Starrise key from the Console
  • Base URL: https://ai.alad.com/v1 (OpenAI-compatible clients)
  • Anthropic-style clients (Claude Code): ANTHROPIC_BASE_URL=https://ai.alad.com/ and ANTHROPIC_API_KEY
Step-by-step guides:

​
Support and errors

Our UAE-based team provides 24/7 support via the Console, email, or WhatsApp.When reporting an issue, include the request ID, timestamp, model, endpoint, HTTP status code, full error message, client and version, relevant configuration (with API keys redacted), and steps to reproduce. Never share a full API key, password, or wallet private key in a ticket, chat, or screenshot.
See Error Handling in the API Reference.