How image generation calls are billed on RelayRouter

RelayRouter's published pricing facts do not list a separate per-image price, so image generation calls are governed by the same per-model rates as other requests. Each model is billed at the live rate shown on relayrouter.io/models, group models settle at a fixed CNY rate per $1 of standard usage, the platform fee is $0 with no minimum spend or subscription, and failed or errored requests are generally not billed.

How is a call priced on RelayRouter?

Each call is priced by the model group it uses, measured against standard usage in US dollars. According to the relayrouter.io/models listing, the GPT group settles at CNY 0.6 per $1 of standard usage and the Claude group at CNY 2.0 per $1, compared with a market reference of CNY 6.8 per $1. Some models use direct pricing instead. For example, deepseek-v4-flash is billed at 1.1x DeepSeek official time-of-day prices: CNY 1.1 per 1M input tokens and CNY 4.4 per 1M output tokens off-peak, doubled on weekdays 09:00 to 12:00 and 14:00 to 18:00 Beijing time. The table below summarizes these published rates.

Which endpoint should an image generation request use?

The provided documentation does not name a dedicated image generation endpoint, so requests go through one of the three documented protocols. RelayRouter is “Compatible with the OpenAI, Anthropic and Gemini protocols”, according to the relayrouter.io site. The documented routes are POST /v1/chat/completions (base https://relayrouter.io/v1), POST /v1/messages (base https://relayrouter.io) and POST /v1beta/models/{model}:generateContent. Authentication uses the header Authorization: Bearer YOUR_API_KEY, with keys created at relayrouter.io/dashboard. Whether a given model returns images, and at what rate, should be confirmed on its listing at relayrouter.io/models and in relayrouter.io/docs before production use.

Do failed image requests or platform fees add cost?

No platform fee applies, and failed calls are generally not charged. According to relayrouter.io, the platform fee is $0, with no minimum spend and no subscription, so the charge for an image generation call comes from the selected model's usage rate. Failed or errored requests are generally not billed, which means a call that returns an error should not appear as usage cost. Payment is made by Stripe card. RelayRouter states that mainstream model groups are on average about 30 percent below official list prices, but the exact figure differs by model and group. For any image workload, read the current per-model rate on relayrouter.io/models rather than relying on the average.

Does switching to RelayRouter change how existing image code is written?

No: existing SDK code needs two configuration changes and nothing else. According to the relayrouter.io docs, you should “Keep your existing SDK, change base_url and the key, no other code changes”. For an OpenAI SDK, set the base URL to https://relayrouter.io/v1; for an Anthropic SDK, use https://relayrouter.io. Streaming is supported. The catalog covers about 108 models across 19 public groups, including the Claude family (claude-opus-5-5, claude-fable-5-1, claude-opus-5), GPT-6 and GPT-5.6 (gpt-6-astra, gpt-5.6-sol), Gemini 3.8 Flash (gemini-3.8-flash), plus DeepSeek, GLM, MiniMax and Moonshot. Because billing is per model, the cost of an image call depends on which model you select.

FAQ

The three questions below cover common billing points for image generation calls on RelayRouter.

Is there a fixed price per generated image?
The provided pricing facts do not list a per-image price. Calls are billed at the selected model's rate, such as CNY 0.6 per $1 for the GPT group or CNY 2.0 per $1 for the Claude group. Check relayrouter.io/models for the live rate.

Am I charged if an image request fails?
Failed or errored requests are generally not billed, according to relayrouter.io.

Do I need a subscription or minimum deposit?
No. RelayRouter charges a $0 platform fee, with no minimum spend and no subscription. Payment is by Stripe card.