Frequently asked questions

Create an account and open the console — a key is issued right there, with no approval step. New accounts also start with a small amount of free trial credit, so you can make your first call before paying anything.

Yes. New accounts start with trial credit, with no card and no KYC required. It is enough to test the API and a few models; when you are ready for more, top up with crypto.

150 frontier models across the families you would expect — chat, reasoning, coding, vision, image, video and audio. New models are added to the gateway as they are verified end to end, and your existing keys gain access to them automatically.

Yes. The same key and the same bill cover text, image, video and audio. Post to /v1/videos with a prompt and wait — a clip takes roughly 40 to 60 seconds and the call returns the finished file with a URL, so give your client a timeout above 60 seconds. Seedance, Kling, Wan and Veo are live today, along with image, speech and music models, and more of the upstream catalogue on request.

Per generation, not per token — the price shown next to each video model on the pricing page is what one clip costs. You only pay for renders that actually complete; failed jobs are not charged.

Yes. The same key and the same base URL serve the OpenAI-compatible format (/v1/chat/completions) and the Anthropic-compatible format (/v1/messages). Point the OpenAI SDK or the Anthropic SDK at the same host and it works.

With crypto: USDT on Binance Smart Chain (BEP-20). Create an invoice in the console, send to the address shown, and your credit is added automatically after on-chain confirmation. No cards, no bank transfers.

No. We do not ask for identity documents, bank details or any approval paperwork. Buy credit and start using it.

Credits are prepaid and unused credit is not refundable. If the service was down on our side, email support and we can adjust your credit for the affected period. Full detail is on the refund policy page.

Blockchain transfers cannot be recalled. If you send the wrong token, on the wrong network, or to the wrong address, those funds are permanently lost and we cannot recover them — so always send only USDT (BEP-20) to the exact address on your invoice.

Yes — each plan carries a requests-per-minute ceiling, from the free tier up to the top plans. Every reseller sub-key can also carry its own budget and its own rate limit, so no single customer can run past its cap.

Yes, on a reseller plan. You get a team account and can issue unlimited customer sub-keys — each with its own budget and rate limit, and each revocable instantly. See the reseller programme.

Yes. If you operate New API, one-api, sub2api, Veloera or done-hub, you can add HELIOROUTE as an upstream channel and keep your own brand in front. See the reseller programme.

Routing fails over where possible, so a wobble at one provider does not have to become your outage. Not every failure can be hidden — if a specific model or provider is unavailable, requests to it may fail until it recovers. Live component status is on the status page.