Flatrouter

Cherry Studio Setup

Connect Cherry Studio to Flatrouter over OpenAI Chat, OpenAI Response, Claude and Gemini.

Cherry Studio is a cross-platform AI desktop client (Windows, macOS, Linux) with multi-model chat, knowledge bases and workflows. It is one of the most capable open-source AI clients right now.

Point it at Flatrouter and you get GPT, Claude, Gemini, DeepSeek and 100+ other models from a single API key.

Four protocols

Cherry Studio can reach Flatrouter over these four protocols:

  • OpenAI Chat — /v1/chat/completions, for GPT-family models such as openai/gpt-4.1 and openai/gpt-5.3-chat
  • OpenAI Response — /v1/responses, for GPT-family models such as openai/gpt-4.1 and openai/gpt-5.4-mini
  • Claude — /anthropic/v1/messages, for Claude-family models such as anthropic/claude-sonnet-4.6 and anthropic/claude-opus-4.6
  • Gemini — /gemini/v1beta/models, for Gemini-family models such as google/gemini-3.1-pro-preview and google/gemini-3.1-flash-lite-preview

Requirements

Setup

Step 1: open settings

Launch Cherry Studio and click the Settings icon in the bottom left.

Step 2: add a provider

Open Model services on the left, scroll to the bottom and click + Add. In the dialog, name the provider and pick its type.

OpenAI Chat — set the provider type to OpenAI, then confirm.

OpenAI Response — set the provider type to OpenAI (turn on Response mode later on the config page), then confirm.

Claude — set the provider type to Anthropic, then confirm.

Gemini — set the provider type to Gemini, then confirm.

Step 3: fill in the API settings

On each provider's config page, enter the API key and API URL.

ProtocolAPI keyAPI URL
OpenAI Chat / ResponseYour Flatrouter API keyhttps://api.flatrouter.com/v1
ClaudeYour Flatrouter API keyhttps://api.flatrouter.com/anthropic
GeminiYour Flatrouter API keyhttps://api.flatrouter.com/gemini

Make sure the toggle in the top right is ON. Otherwise the provider will not appear in the model picker.

Step 4: add models

Click Manage in the model area to pull the list, then click + next to a model to enable it.

Step 5: verify

Click Detect, pick any model, and run the test. A pass means the setup worked.

Start chatting

Back on the main screen, open the model picker at the top of the chat and choose a model under the matching provider group.

FAQ

Q: The model list is empty after I click "Manage"

Check that the API URL is correct (no trailing slash) and that the API key was copied in full, with no extra spaces.

Q: Detect reports a connection failure

  1. Copy the API key again from the Flatrouter console, with no extra spaces
  2. Check the API URL
  3. Check your network

Q: The provider does not appear in the model picker

Check that the toggle in the top right of the config page is on.

On this page