Cherry Studio

A fuller-featured desktop client, suited to managing several models and multiple assistants.

OS	Package
Windows (x64 / ARM64)	`setup.exe` installer, or the `portable.exe` no-install build
macOS (Intel / Apple Silicon)	Pick the `.dmg` matching your processor; if the first launch is blocked, allow it under Privacy & Security
Linux (x86_64 / ARM64)	`.deb` for Debian/Ubuntu, `.rpm` for Fedora/RHEL, `.AppImage` for everything else

Before you start

You need an API Key with available quota. Copy model names from the model list — they are case-sensitive.

Configuration

The steps are the same on all three platforms.

After launch, click Choose another provider
Click Add in the bottom-left
Fill in:

Field What to enter
Provider name TokenRoute (your choice)
Provider type OpenAI
API Key Your API Key
API address https://api.smartwan.com
Click Fetch model list
Click Add all models on the right, or pick the ones you want manually
Return to Home and select a TokenRoute model in the default assistant's model picker

Cherry Studio detects each model's capabilities automatically; when it gets one wrong, click ⚙️ to adjust manually.

With a model selected on the home screen, send:

Hello, please reply: Cherry Studio setup successful

Symptom	Cause
Fetching the model list fails	Wrong API address or key; confirm the address is `https://api.smartwan.com`, without `/v1`
Models added but chat errors out	Check account quota, and whether that model is within this key's allowed set
Capabilities detected incorrectly	Click ⚙️ to enable image, tool and other capabilities by hand