What it costs

There is no gateway rate card to learn. Whatever OpenAI, Anthropic, Google or DeepSeek charges for the model you called is multiplied by 1.2 and converted to Toman. Both inputs are public (the provider's published price, and the token counts returned in the response), so you can check any line on your bill yourself.

# what you are billed for one call
cost_usd   = provider_list_price(model, prompt_tokens, completion_tokens)
billed_usd = cost_usd * 1.20          # flat 20%, every model, every endpoint
billed_toman = to_toman(billed_usd)   # deducted from your wallet balance

# the token counts in that formula are the ones the API hands back
resp = client.chat.completions.create(...)
resp.usage.prompt_tokens, resp.usage.completion_tokens

Live per-model rates in Toman are on the pricing page and the models page; both read from the same table the biller uses, which is why they are not reprinted here. To size a workload before running it, the cost calculator and token counter take a prompt and a model and return the Toman figure.

API · pay as you go

Deducted per call from wallet credit at list price plus 20%. No plan required, no monthly minimum, no expiry on credit you bought.

Chat · free

A 100,000-Toman allowance every 30 days, auto-renewing, with a 40,000-Toman weekly cap, on gpt-4o-mini, claude-haiku-4-5 and gemini-2.5-flash.

Chat · Base

500,000 Toman per month.

Chat · Pro

2,000,000 Toman per month.

Chat · Max

10,000,000 Toman per month.

Top-ups

Zarinpal, minimum 100,000 Toman. A local card is all that is needed at any point in the flow.

Per-plan allowances, weekly caps and model ranges are listed on the plans page. The hosted chat is a web app that installs as a PWA on phone or desktop; there is no native app and 1xAi is not in any app store.

1. [01]

   The free chat plan. 100,000 Toman of chat allowance every 30 days, renewing automatically, capped at 40,000 Toman per week, on the three fast models. Signing up needs an email — no phone number and no foreign card.

2. [02]

   The signup gift wheel. One server-drawn spin per new account, worth from 100,000 to 1,000,000 Toman in wallet credit. It never expires, and unlike the chat allowance it can be spent on API calls, which makes it the straightforward way to benchmark the endpoint against your own workload.

Provider list prices are the providers' to change, and so is the Toman conversion; the 20% rule is what stays fixed. Credit bought with money does not expire, but plan allowances are periodic by definition: they reset rather than accumulate. And a gateway cannot promise a provider's behaviour: model availability and usage policy remain theirs.

1xAi is an independent gateway. It is not affiliated with, endorsed by, or a reseller for OpenAI, Anthropic, Google or DeepSeek. Requests are relayed to those providers' official APIs, so each provider's own usage policies apply to what their models will and will not do, and their model availability is theirs to change. Current reachability is published on the status page.