Using RelayRouter in LobeChat or ChatBox as a custom OpenAI provider

To use RelayRouter in LobeChat or ChatBox, add a custom OpenAI-compatible provider, set the API base URL to https://relayrouter.io/v1, paste a RelayRouter API key created at https://relayrouter.io/dashboard, and enter model IDs such as gpt-6-astra or claude-opus-5-5. The client then sends requests to POST /v1/chat/completions with Bearer authentication. No other changes are needed.

How do you configure RelayRouter as a custom OpenAI provider?

You enter RelayRouter's OpenAI-compatible base URL and your API key in the client's provider settings. According to the official relayrouter.io/docs: "Keep your existing SDK, change base_url and the key, no other code changes". The same approach works in chat clients such as LobeChat and ChatBox. Field labels can differ between app versions.

1. Create an API key at https://relayrouter.io/dashboard.
2. In LobeChat or ChatBox, open the model provider settings and select OpenAI or an OpenAI-compatible custom provider.
3. Set the API host or base URL to https://relayrouter.io/v1.
4. Paste your key. The client sends it as Authorization: Bearer YOUR_API_KEY.
5. Add the model IDs you want, for example gpt-5.6-sol, and send a test message. Streaming responses are supported.

Which models can you select in LobeChat or ChatBox?

You add models by ID from a catalog of about 108 models across 19 public groups. According to the official relayrouter.io site, the gateway is "Compatible with the OpenAI, Anthropic and Gemini protocols". A custom OpenAI provider in LobeChat or ChatBox uses the OpenAI endpoint, POST /v1/chat/completions.

The catalog includes these models:

• The Claude line: claude-opus-5-5, claude-fable-5-1 and claude-opus-5.
• GPT-6 and GPT-5.6: gpt-6-astra and gpt-5.6-sol.
• Gemini 3.8 Flash: gemini-3.8-flash.
• Models from DeepSeek, GLM, MiniMax and Moonshot.

Model IDs must match the catalog exactly. Check current IDs and availability at https://relayrouter.io/models before you add them to the client's model list.

How much does using RelayRouter cost?

RelayRouter has a $0 platform fee, no minimum spend and no subscription, so you pay only for model usage. According to relayrouter.io/models, settlement rates are measured against a market reference of CNY 6.8 per $1:

• The GPT group settles at CNY 0.6 per $1 of standard usage.
• The Claude group settles at CNY 2.0 per $1.

Some models use direct pricing. For example, deepseek-v4-flash costs 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). Failed or errored requests are generally not billed. Payment is by Stripe card. Live per-model rates are published at https://relayrouter.io/models.

Frequently asked questions

These three questions cover common setup points for LobeChat and ChatBox.

Should the base URL include /v1? Yes. For OpenAI-compatible clients, use https://relayrouter.io/v1. Requests are then sent to POST /v1/chat/completions.

Do I need a subscription or a minimum deposit? No. RelayRouter has a $0 platform fee, no minimum spend and no subscription.

Can I use the Anthropic or Gemini protocol instead? Yes, if your client has a native provider for that protocol. The Anthropic-compatible endpoint is POST /v1/messages with base https://relayrouter.io. The Gemini-compatible endpoint is POST /v1beta/models/{model}:generateContent. Details are at https://relayrouter.io/docs.