Novin Cloud AI

Novin Cloud AI is a single gateway for accessing dozens of large language models. Instead of creating a separate account, obtaining a key, and paying a foreign-currency bill for every provider, you get access to all models with one API key and Rial payments from your Novin Cloud wallet.

Why use this service?

Without Novin Cloud	With Novin Cloud
A separate account for every provider	One user account
A separate API key for every service	One key for all models
Requires an international credit card	Rial payment from your wallet
Access issues from Iran	Direct access, no intermediary
Different code for every provider	One standard, OpenAI-compatible API

Full OpenAI compatibility

The most important feature of this service is full compatibility with the OpenAI API standard. This means any library, framework, or tool that works with OpenAI connects to Novin Cloud by changing just two values:

from openai import OpenAI

client = OpenAI(
    api_key="your-API-key",                 # Your Novin Cloud key
    base_url="https://iapi.novin.cloud/v1"  # Novin Cloud endpoint
)

That's it. Without changing another line of code, you now have access to every model. This means LangChain, LlamaIndex, Cursor, Continue, and any other tool that supports OpenAI works without needing a special plugin.

Available models

Models from several providers are available:

• Anthropic — the Claude family (Opus, Sonnet, Haiku)
• OpenAI — the GPT family and the o-series reasoning models
• Google — the Gemini family, with a very large context window
• Meta — the Llama family
• DeepSeek — chat and reasoning models
• Mistral — including Codestral for coding
• Qwen — including Qwen Coder

The full, up-to-date list of models, with pricing and specifications, is available on the Models page and in the Models section of the user console.

tip

The model list is read live from the gateway. For the most accurate status, check the user console or use the standard /v1/models endpoint.

How the service works

       You / your application
               │
               │  Authorization: Bearer sk-...
               ▼
    ┌──────────────────────┐
    │  iapi.novin.cloud    │   AI gateway
    └──────────┬───────────┘
               │
               ▼
        AI model providers

Gateway — All requests pass through iapi.novin.cloud. This layer routes each request to the relevant provider, enforces your key's limits, and counts the tokens consumed.

Authentication — API keys are tied to your user account and are created and managed from the user console.

Cost calculation — The cost of each request is calculated from the input and output tokens and deducted from your Rial wallet. See details on the Usage & Billing page.

Two ways to use it

1. Through the API

For use in applications, scripts, and development tools. Create an API key and use it:

curl https://iapi.novin.cloud/v1/chat/completions \
  -H "Authorization: Bearer $NOVIN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Full guides: Quickstart and API Reference.

2. Through the chat interface

If you want to work with the models without writing code, you have two options:

• In-console chat — the Chat section of the user console, for quick conversations and trying out models. Guide: Chat
• Dedicated chat interface — chat.novin.cloud, with more complete features, using the same Novin Cloud account to sign in

How billing works

Costs are calculated based on actual usage — there is no monthly subscription or fixed fee. You only pay for the tokens you actually consume, and the cost is deducted from your Rial wallet balance.

Each model's price is calculated separately based on the number of input tokens and the number of output tokens. Output tokens are usually more expensive than input tokens.

Insufficient balance

If your wallet balance is insufficient, requests are rejected with a 402 error. Make sure your wallet is topped up before use.

Next steps

• 🚀 Quickstart — from creating a key to your first request, in a few minutes
• 🔑 API Keys — creating keys, setting budgets and rate limits
• 🧠 Models — choosing the right model for your task
• 💬 Chat — using the console chat interface
• 📊 Usage & Billing — tracking costs
• 📖 API Reference — complete endpoint documentation