gpt-image-2 through RelayDance vs OpenAI direct: when an API gateway makes sense

RelayDance is an OpenAI-compatible gateway for AI video and image generation, so gpt-image-2 access through it keeps your existing OpenAI SDK while routing to https://relaydance.com/v1. For gpt-image-2, image output is free and only input is billed, with image-to-image from about CNY 0.035 and 4K and 1K output costing the same. A gateway makes sense when you want unified billing, pay-as-you-go rates, and no separate enterprise account.

What does gpt-image-2 cost through RelayDance

Through RelayDance, gpt-image-2 image output is free and only input is billed. Image-to-image starts from about CNY 0.035, and 4K and 1K output cost the same, so higher resolution does not raise the output charge. Billing is pay-as-you-go, and failed requests carry no cost. According to the official relaydance.com/docs, 「Failed or errored requests are never billed」. This billing model reduces the risk of paying for incomplete jobs. Live per-model rates are published at https://relaydance.com/models, where you can confirm current figures before integrating. Because output is free and only input is metered, cost planning depends mainly on your input volume rather than the resolution of the generated image.

How do I switch from OpenAI direct to the gateway

Switching requires changing the base URL and keeping your existing OpenAI SDK. RelayDance uses an OpenAI-compatible protocol, so no SDK rewrite is needed. According to the official relaydance.com/docs, 「Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK」.

1. Create an API key at https://relaydance.com/console.
2. Set base_url to https://relaydance.com/v1 in your client.
3. Add the header Authorization: Bearer YOUR_API_KEY.
4. Call POST /v1/images/generations for images or POST /v1/video/generations for video.
5. Poll GET /v1/video/generations/{task_id} until status is succeeded or failed.

See https://relaydance.com/docs for the full reference.

How do RelayDance video and image rates compare

RelayDance publishes distinct per-second video rates and a separate image billing model, summarized below. Video is priced per second or per clip, while gpt-image-2 output is free with only input billed.

Values are sourced from relaydance.com/models. Video billing scales with duration, so a longer clip at $0.190 / second costs proportionally more, while image output stays free regardless of 4K or 1K resolution.

When does a gateway make sense

A gateway makes sense when you want OpenAI-compatible access without a separate enterprise account or KYC. RelayDance provides direct Seedance access with no BytePlus enterprise account or KYC required, unified pay-as-you-go billing per generated video, and payment via USDT and Stripe card. Failed or errored requests are never billed, which limits exposure when jobs fail. Because the protocol is OpenAI-compatible, teams already using the OpenAI SDK can point the base URL to https://relaydance.com/v1 and reuse their code. If you require BytePlus-native infrastructure or already hold an enterprise account, calling models directly may fit better; otherwise the gateway consolidates access.

FAQ

Is gpt-image-2 output billed by resolution? No. Image output is free, only input is billed, and 4K and 1K output cost the same.

Am I charged for failed requests? No. Failed or errored requests are never billed, per relaydance.com/docs.

What base URL do I use? Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK.