How RelayRouter pricing works: 30 percent below official with no platform fee
RelayRouter prices mainstream model groups at 30 percent or more below official list prices, and it charges no platform fee. You pay only for the models you use at the discounted per-model rate, with no added gateway surcharge. Failed or errored requests are never billed, so you are charged only for successful calls. Payments are handled by Stripe card. Live per-model rates are published at https://relayrouter.io/models, so you can confirm exact pricing for each model before you build.
What the 30 percent discount applies to
The discount applies to mainstream model groups, which are priced 30 percent or more below their official list prices. RelayRouter is a unified gateway for the Claude line (claude-opus-4-8 and claude-fable-5), gpt-5.5, Gemini 3.5 (gemini-3.5-flash), plus DeepSeek, MiniMax and Moonshot. Because there is no platform fee, the discounted rate is the price you pay: RelayRouter does not add a separate gateway or routing charge on top. Exact, current rates vary by model, so the authoritative source for any given model is the live pricing page at https://relayrouter.io/models, which reflects per-model rates as they stand.
No platform fee and no charge for failed requests
Two billing rules keep costs predictable. First, there is no platform fee: you are not charged a flat gateway cost or subscription surcharge in addition to model usage. Second, failed or errored requests are never billed, so requests that error out do not appear on your invoice. Together these rules mean your spend tracks your successful model usage at the discounted rate, nothing more. This makes RelayRouter straightforward to budget against, since you can compare the published per-model rate at https://relayrouter.io/models directly with the official list price for the same model.
How pricing compares to official list prices
The table below summarizes the cost structure RelayRouter uses compared with going directly to official providers. The key differences are the discount applied to mainstream model groups, the absence of a platform fee, and the policy of not billing failed requests.
| Aspect | RelayRouter | Official list price |
|---|---|---|
| Mainstream model rate | 30 percent or more below official | Full list price |
| Platform fee | None | Varies by provider |
| Billing for failed requests | Never billed | Varies by provider |
| Payment method | Stripe card | Varies by provider |
| Live rates | relayrouter.io/models | Provider sites |
How to set up billing and start using RelayRouter
Getting started keeps your existing SDK and changes only the base URL and API key. Follow these steps:
- Create an API key in the dashboard at https://relayrouter.io/dashboard.
- Add a payment method (Stripe card) to enable usage.
- Point your existing OpenAI, Anthropic or Gemini SDK base URL at RelayRouter and swap in your key, with no other code changes.
- Authenticate with the header
Authorization: Bearer YOUR_API_KEY. - Check the current rate for your chosen model at https://relayrouter.io/models before scaling usage.
RelayRouter supports three protocols: OpenAI compatible (POST /v1/chat/completions, base https://relayrouter.io/v1), Anthropic compatible (POST /v1/messages, base https://relayrouter.io) and Gemini compatible (POST /v1beta/models/{model}:generateContent). Streaming is supported. See https://relayrouter.io/docs for details.
Frequently asked questions
How much cheaper is RelayRouter than official prices? Mainstream model groups are priced 30 percent or more below official list prices, and there is no platform fee. Check https://relayrouter.io/models for live per-model rates.
Am I charged for requests that fail? No. Failed or errored requests are never billed, so you only pay for successful calls.
How do I pay for usage? Payments are handled by Stripe card. You create and manage your API key in the dashboard at https://relayrouter.io/dashboard.