Skip to main content

Cherry Studio

Cherry Studio is a cross-platform AI desktop client that supports multiple LLM providers. By connecting it to ShuYou, you can use ShuYou’s aggregated model catalog in Cherry Studio with flexible model selection and unified billing.
ShuYou fully supports the OpenAI Chat Completions API. Add ShuYou as a custom OpenAI-compatible provider in Cherry Studio to use all models labeled OpenAI API Compatible.OpenAI protocol base URL: https://api.shuyou.ai/v1

Configuration

Step 1: Download and install Cherry Studio

Download the client for your OS from the Cherry Studio download page and complete installation.

Step 2: Add the ShuYou model provider

  1. Open Cherry Studio and go to Settings → Model Provider.
  2. Click Add to create a new provider.
  3. Fill in the configuration dialog:
FieldValue
Provider NameShuYou (or any name you prefer)
Provider TypeOpenAI
API KeyYour ShuYou API Key
API Hosthttps://api.shuyou.ai/v1/
Replace the placeholder with your real ShuYou API Key from the console. The API Host must end with a trailing slash (/).
Configuration notes:
  • Provider Name — Customizable; ShuYou is recommended for clarity.
  • Provider Type — Must be OpenAI; ShuYou exposes models via the OpenAI Chat Completions protocol.
  • API Key — Used for authentication and billing on ShuYou.
  • API Host — ShuYou’s OpenAI-compatible endpoint.

Step 3: Fetch the available model list

After saving the provider, click Manager. Cherry Studio will call ShuYou’s /v1/models endpoint and list all OpenAI-compatible models—no manual slug entry required.

Step 4: Add the models you need

From the fetched list, select models to add to your library. Model recommendations (adjust to your workload and budget):
  • High performance: openai/gpt-5, anthropic/claude-sonnet-4.5
  • Balanced: google/gemini-2.5-pro, x-ai/grok-4-fast
  • Cost-effective: deepseek/deepseek-chat, qwen/qwen3-coder-plus
See the full catalog on the ShuYou model list.

Step 5: Start a conversation

Open the chat interface, select a model under the ShuYou provider, and send a message.
You can switch models mid-conversation. Cherry Studio keeps chat history, which makes it easy to compare outputs across models.

Supported models

ShuYou offers a large and growing set of OpenAI API Compatible models. To explore them:
  • Filter OpenAI API Compatible on the ShuYou model list
  • Or check protocol support on each model’s detail page

Usage experience

Once configured, you can use ShuYou models in Cherry Studio with:
  • A unified chat UI without switching clients
  • Real-time model switching
  • Chat history management and export
  • File uploads and multimodal input (when the selected model supports them)
Capabilities such as vision and function calling vary by model. Check each model’s detail page before relying on a specific feature.

Troubleshooting

API Key error

Issue: Invalid or unauthorized API Key when adding the provider. Solutions:
  • Copy the key carefully—avoid extra spaces or line breaks.
  • Confirm the key is active and your account has sufficient balance.
  • Verify the key in the ShuYou console.

Unable to fetch model list

Issue: No models appear after clicking Manager. Solutions:
  • Set API Host to https://api.shuyou.ai/v1/ (with trailing slash).
  • Check network access to https://api.shuyou.ai.
  • Ensure firewall or proxy settings are not blocking requests.
  • Update Cherry Studio to the latest version.

Model invocation failed

Issue: Errors after selecting a model and sending a message. Solutions:
  • Confirm the model is still available on ShuYou.
  • Check account balance and plan limits.
  • Verify request parameters (temperature, max_tokens, etc.) are supported by the model.
  • Read the error message and adjust parameters.
  • Try another model to isolate provider-specific issues.

Slow response

Issue: Long latency or stalled replies. Solutions:
  • Check network quality.
  • Try a faster model slug (for example x-ai/grok-4-fast).
  • Lower max_tokens to reduce generation time.
  • Retry during off-peak hours or choose a less loaded model.