Common questions

A single access point to AI models: one API key and one balance unlock GPT, Claude, Gemini, DeepSeek and dozens more. The API is OpenAI-compatible, so nearly any existing SDK or tool works out of the box.

Sign up, top up your balance and create an API key in the dashboard. Then paste the key into a snippet from the docs and send your first request — the whole thing usually takes under ten minutes.

Over 70 models from 19 providers: text, images, video, music and embeddings. The full list with prices is on the Models & pricing page; the dashboard has a catalog with filters and favorites.

No. The service works from Russia without a VPN; top up with a Russian card or SBP. For euro payments there is a separate option — a card via Stripe.

In the dashboard, under Billing: with a Russian card or SBP in rubles, or with a card via Stripe in euros. Funds are credited as soon as the payment is confirmed.

Text models are charged by input and output tokens at the model price at request time. Media models have a fixed price per generation. Every charge is visible in the transaction history and stats.

Requests start returning a 402 error — no penalties, no account locks. Top up and requests resume immediately; nothing needs reconfiguring.

Your choice: rubles, dollars or euros — the switcher is in the site header. Internal accounting is in dollars; display prices are converted at the current rate.

Yes. In Billing, enable balance notifications: set a threshold and an email — we send a message when the balance drops below it.

Yes. Share your link from the Referrals section — 5% of every top-up made by people you invited lands straight on your balance.

Point the official SDK base_url at our service with the /v1 suffix and use your ya-… key. The rest of your code stays the same. Ready-made snippets for curl, Python and JavaScript are in the docs.

Change the model parameter in the request — it goes to a different provider. Your code, SDK and key stay the same.

Yes. Pass "stream": true — the response arrives in SSE chunks, just like OpenAI, and ends with data: [DONE].

Yes, into any tool that supports an OpenAI-compatible API: Cursor, Claude Code, Cline, n8n, LangChain and more. See the Tools section for the collection.

No separate integration is needed: YourAPI itself is a single access point with an OpenAI-compatible API — connect directly with one key.

We never store full keys — only their hash, so showing a key again is technically impossible. Copy it at creation; if lost, create a new one and revoke the old.

On the key page, set a spend limit per day, week or month. Once reached, the gateway returns 429 and stops charging until the next window starts.

Yes. Every key has an access schedule: mark the allowed hours per weekday in your time zone — outside those hours requests are rejected.

Pause or revoke it in the dashboard — the block takes effect within a minute. The key's request history stays in your stats.

Yes, per key and per IP. Exceeding it returns 429 with a Retry-After header — retry after the indicated pause.

On the key page: spend and token charts, a 24-hour availability bar and live logs in real time. Logs can be downloaded as CSV or run through AI analysis.

No. Request and response bodies are never stored — not in the database, logs or analytics. Stats keep metadata only: model, tokens, cost, time.

Yes, as CSV: request stats from the API keys page, transaction history from Billing. The files open in Excel and Google Sheets.

The Server status page shows hourly availability of the gateway and models for the last 24 hours. If your requests are failing, check there first.

No. Payments are processed by the payment providers on their side — we neither receive nor store your card details.