Frequently asked questions
Keys, balance, payments and what happens to your requests.
25 questions in 5 sections
The basics
What exactly is Evrona?
One way in to AI models: a single API key and a single balance unlock GPT, Claude, Gemini, DeepSeek and dozens of other models. The API is OpenAI-compatible, so almost any existing SDK or tool will fit.
Where do I begin?
Sign up, put money on your balance and issue an API key in the dashboard. Then paste the key into a snippet from the docs and send the first request — all of it usually takes less than ten minutes.
What models does the catalog hold?
More than 70 models from 19 providers: text, images, video and embeddings. The complete list with prices is on the Catalog & prices page, and the dashboard has a catalog with filters and favorites.
Will I need a VPN or a foreign card?
No. The service is reachable from Russia without a VPN, and a foreign card is not required: the balance is topped up with a code — buy it in the store, activate it in the dashboard, and the amount is credited immediately.
Balance & payments
What are the ways to top up my balance?
With a top-up code: buy the code in the store, open Billing in your dashboard and activate it — the amount lands on your balance straight away. Card and SBP payment directly on the site will appear later.
By what rule is the cost of a request charged?
Text models are billed by input and output tokens at the model price at the moment of the request. Media models carry a fixed price per generation. Every charge shows up in the transaction history and in the stats.
What happens once the balance hits zero?
Requests will start returning a 402 error — no penalties and no account lock. Top up the balance and requests continue at once; nothing has to be reconfigured.
Which currencies can prices be shown in?
It is up to you: rubles, dollars or euros — the switcher sits in the site header. Internal accounting runs in dollars, and displayed prices are recalculated at the current rate.
Will I be warned about a low balance?
Yes. Open Billing and switch on balance notifications: set a threshold and an email — as soon as the balance falls below the threshold, a letter arrives.
Do you run a referral program?
Yes. Send out your link from the Referrals section — 5% of each top-up by an invited user comes straight to your balance.
API & integration
What does it take to connect through an OpenAI-compatible SDK?
Point the official SDK base_url at our service with the /v1 suffix and use your ev-… key. The rest of your code stays the same. Ready-made snippets for curl, Python and JavaScript are in the docs.
How do I move from one model to another?
Change the value of the model parameter in the request — and it will be sent to a different provider. Your code, SDK and key remain unchanged.
Is response streaming available?
Yes. Send "stream": true — the response will come as SSE chunks, the same as OpenAI, and finish with data: [DONE].
Will Evrona work inside the tools I already use?
Yes, to a wide range of ready-made integrations: Cursor, Claude Code, Cline, n8n, LangChain and others. The collection is gathered in the Tools section.
Does Evrona work with OpenRouter?
You don't need a middleman: Evrona already gives you a single access point and an OpenAI-compatible API, so you connect directly with one key.
Keys & restrictions
Why does the API key appear just once?
We do not keep full keys — only their hash, so showing a key a second time is technically impossible. Copy it right at creation; if it is lost, create a new one and revoke the previous one.
How do I put a ceiling on a key's spending?
On the key page, set a money limit for a day, a week or a month. As soon as the limit is reached, the gateway returns 429 and stops charging until the next window begins.
Can I give a key working hours?
Yes. Each key has an access schedule: mark the permitted hours for every weekday in your own time zone — beyond them requests are turned down.
What should I do if a key leaks?
Pause the key or revoke it in the dashboard — the block comes into force within a minute. The request history for that key stays in the stats.
Is a request rate limit in force?
Yes, a rate limit applies per key and per IP. On exceeding it you get 429 with a Retry-After header — repeat the request after the stated pause.
Where do I watch what a key is doing?
On the key page: spend and token charts, a 24-hour availability bar and live logs in real time. The logs can be exported as CSV or sent for AI analysis.
Reliability & data
Are my prompts and the models' responses stored?
No. Neither the database, nor the logs, nor analytics keep request and response bodies. Only metadata remains in the stats: model, tokens, cost, time.
Is it possible to export stats and transaction history?
Yes, in CSV: request stats are exported from the API keys page, transaction history from Billing. Both files can be read in Excel and Google Sheets.
Where do I see service outages?
The Service uptime page shows hourly availability of the gateway and the models over the past 24 hours. If your requests have started failing, look there first.
Do my payment details travel through your service?
No. Payments are processed by the payment providers on their own side — your card details never reach us and are not stored.
Didn’t find an answer?
Write to us — we will reply by email. support@evrona.ru