Use Cherry Studio with llmrelay

Setup

1. 1 Click Settings (gear icon) in the left navigation bar, then open the Model Services tab.
2. 2 Click + Add below the provider list, name the provider LLM Relay and set Provider type to Anthropic (choose OpenAI if you prefer the /v1 Chat Completions path).
3. 3 Click Add, then open the new provider in the list and paste your key into API key.
4. 4 Set API address (Base URL) to https://api.llmrelay.dev for the Anthropic type, or https://api.llmrelay.dev/v1 for the OpenAI type.
5. 5 Click + Add under model management and enter model IDs such as claude-sonnet-5, or use Manage to fetch the list.
6. 6 Toggle the provider's enable switch on the right of the list.

Config

# Cherry Studio: Settings -> Model Services -> + Add
# Provider name: LLM Relay
# Provider type: Anthropic        (OpenAI / Gemini / Anthropic / Azure OpenAI are the options)
# API key:       ***
# API address:   https://api.llmrelay.dev      # OpenAI type would use https://api.llmrelay.dev/v1
# Models:        claude-sonnet-5, claude-opus-5, gpt-5.6-luna

Smoke test

Copy, paste, run. If you get JSON back with a message, you're good.

Click the Check button next to the API key field; it must report the key as valid, then start a chat with one of the added models.

Gotchas

• !Custom provider type is limited to OpenAI, Gemini, Anthropic and Azure OpenAI; there is no free-form adapter.
• !The docs warn to keep http and https straight and to paste only the host (e.g. https://xxx.com) without a trailing slash or extra path when copying a base URL from a gateway console.
• !Cherry Agent requires an Anthropic-protocol endpoint, so use the Anthropic provider type if you want the relay to back agents rather than plain chat.

Related integrations