Skip to main content
Cherry Studio is a graphical chat client that supports multiple providers side by side. This page walks through the full setup for Global AI, using Claude as the example. The steps for OpenAI and Gemini are identical.
Before you begin, complete the redemption and key steps in the Integration overview.
Download: cherry-ai.com

​
Configuration

1

Open Model Services

Click the gear icon in the top right, then open Model Services on the left.
2

Configure the API URL and key

At the top of the provider list, search for New-API and select it. In the New-API form on the right:
  • API Key: paste your sk- token, see Token management for details.
  • API URL: https://globalai.vip
Check for stray whitespace at the start and end when pasting; otherwise the connection will fail.
3

Fetch the model list

Click Get model list and Cherry will fetch every available model automatically. Search for the models you need (for example claude-sonnet-5, gpt-5.6-sol), click + next to each one to add it to the active list, then close the settings dialog.
4

Start chatting

Click Home in the top left, open the chat view, choose the model you added from the top dropdown, and start using it.

​
Troubleshooting

The key was copied incompletely or with whitespace. Recopy from /keys and check.
The most common cause is a proxy or circumvention tool. Turn them off and retry.
The model ID is misspelled, or the model is not allowed in your token’s scope. Search for it on the Pricing page to confirm the spelling.

Related topics

Global AI Client Integrations: Parameters and SelectionChatBox with Global AI: Desktop and Mobile SetupChatWise Setup: Unified Multi-Provider Access via Global AIGlobal AI Quickstart: From Sign-Up to First API CallGlobal AI Token Management: Create, Limit, and Secure