Billing & Pricing
Nilux AI uses a pay-per-use billing model. There are no subscriptions, no monthly fees, and no hidden charges. You pay only for the tokens your requests consume.
How Billing Works
- Top up your balance — Add funds to your account (self-service top-up coming soon)
- Use Nilux AI — Each request consumes tokens, measured in real time
- Cost is deducted — After each request, the exact token cost is subtracted from your balance
- Never overspent — When your balance reaches $0, requests are blocked. You never go into debt
Pricing Table
Prices are per 1 million tokens. These are the final prices you pay — infrastructure and service costs are already included.
| Model | Input (per 1M) | Output (per 1M) | Cached Input (per 1M) |
|---|---|---|---|
| Standard | $0.21 | $0.42 | $0.0042 |
| Pro | $0.65 | $1.31 | $0.0054 |
Tip: Standard is the recommended model for most tasks. Switch to Pro when you need deeper reasoning.
Smart Caching
Nilux AI uses intelligent caching to significantly reduce your costs. After the first request in a session, the system prompt and conversation context are cached. Subsequent requests reuse the cache — cached input tokens cost up to 50x less than fresh tokens.
In typical coding sessions, 85-98% of input tokens are cached, which means your actual costs are much lower than the base input price.
Example: A 10M token session on Standard with 90% cache rate costs approximately $1.40 instead of $2.10.
Checking Your Balance
Use the /balance slash command inside the CLI:
> /balance
Your balance: $4.52
You can also view your balance and usage statistics in the Dashboard.
When Your Balance Runs Out
If your balance reaches $0:
- New requests are blocked with a clear message
- No debt is incurred — you only pay for what you consume
- Top up your balance to resume
Top Up Your Balance
Currently, balance top-ups are handled via support. Contact us to add funds:
- Email: nilline2017@gmail.com
Self-service top-up is coming soon. You will be able to add funds directly from the Dashboard.
Usage Statistics
The Dashboard provides detailed usage breakdowns:
- Total tokens consumed per period (7, 30, or 90 days)
- Cost per task type (chat, subagent, web search)
- Average tokens per request
Access this at Dashboard > Usage.
FAQ
How is token count measured?
Token counting uses the same tokenizer as the underlying AI provider. Both input tokens (your conversation context) and output tokens (the model's response) are counted.
Do unused credits expire?
No. Your balance never expires. Credits remain in your account until consumed.
What if a request fails mid-response?
You are only charged for tokens actually processed. If the connection drops or the provider returns an error, you are not charged for incomplete responses.