CupAi AI Model Pricing & Packages
| Model | Input (Toman / million tokens) | Output (Toman / million tokens) | Cache (Toman / million tokens) | Cache Write (Toman / million tokens) |
|---|
At CupAi, you can purchase credit packages to access advanced AI models like GPT, Claude, and Gemini. Costs are calculated based on the number of tokens consumed, and you only pay for actual usage—enjoy content generation at the original provider’s prices with no markup.
Our packages are designed according to file size, processing time (within 3 or 24 hours), and the model type. You can experience photo editing, translation, and content generation at the original model prices without any additional cost.
How to choose the right model
Every model strikes a different balance between speed, answer quality and cost. Larger models such as GPT and Claude suit complex reasoning, long-document analysis and precise coding, while lighter models cost far less for summarising, classification and short replies. The table above lists each model's price split by input and output tokens, so you can estimate a job's cost before you start.
A token is the smallest unit of text a model processes: roughly 750 English words, or about 500 Persian words, per thousand tokens. The cost of a request is the sum of its input and output tokens, so shortening your prompt and capping the response length reduces the bill directly. For a closer estimate, use the calculator on this page.
Frequently asked questions about model pricing
- How is the cost calculated?
- By the number of input and output tokens in each request. Input and output rates are published per model and shown in the table on this page.
- Do prices differ from the provider's own?
- The base rate is the provider's own rate and CupAI adds no percentage on top; only a few models carry a usage multiplier, which is shown in the same table.
- Which model should I start with?
- A mid-tier model is enough for everyday writing and summarising. Reach for a flagship model when the output needs high accuracy or the input text is very long.
- How do payments work?
- All payments are in local currency and no foreign credit card is needed. Credit you buy can be spent on any model available to you.
| Model | Model provider | Input price (thousand tokens / Toman) | Output price (thousand tokens / Toman) | Cache Ratio | Cache Write | Total (Toman) |
|---|