Skip to content
NoviaHub Docs

Desktop apps

Cherry Studio

Add NoviaHub to Cherry Studio and chat with NoviaHub models from a desktop client.

Cherry Studio is a desktop AI client for Windows, macOS and Linux. It has a built-in provider called “New API”, and NoviaHub is compatible with it: enter NoviaHub’s address and your API key, and that one provider can use the OpenAI Chat, OpenAI Responses, Anthropic Messages and Gemini protocols as needed.

This page follows Cherry Studio v2 (v2.1.3, released 2026-09-24). Labels are quoted from the app’s English interface.

Prerequisites

Section titled “Prerequisites”
  • A NoviaHub API key. If you don’t have one yet, see API keys.
  • Funds in your account. Calls are charged to your balance by usage.

Install

Section titled “Install”

Download the package for your system from Cherry Studio’s official download page, cherryai.com.cn/download/v2:

System Packages (v2.1.3)
Windows (x64, arm64) Installer .exe or portable .exe
macOS (Apple silicon, Intel) .dmg
Linux (x64, arm64) AppImage, .deb, .rpm

Configure NoviaHub

Section titled “Configure NoviaHub”

Use either method. The one-click import saves you typing the address and key, but with both methods you still have to add models and turn on the enable switch.

  1. Make sure Cherry Studio is installed on this computer.

  2. Sign in to NoviaHub and open the API Keys page. Find the key you want to use, click ⋯ at the right end of its row, then choose Chat → Cherry Studio.

  3. Your browser asks whether to open Cherry Studio; allow it. NoviaHub passes its address https://noviahub.com and this key to Cherry Studio’s built-in New API provider through a cherrystudio:// link.

  4. Cherry Studio shows a confirmation dialog titled “Add New API API Key”, listing the Provider Name, the Provider ID (new-api) and the Base URL (https://noviahub.com). Check them and click Add.

    • “New API already has this API key. It will not be added again.” means this key was imported before.
    • If you see “Failed to add provider API key, data format error”, configure it by hand with Method 2 instead.

  5. The import only writes the address and the key. It does not add models and does not turn on the provider’s enable switch. Continue with steps 4–6 of Method 2 to add models, check the connection and enable the provider.

Verify

Section titled “Verify”
  1. Under Model Provider, click Check, pick a model and make sure the check passes.
  2. Start a new chat, pick a model under New API in the model selector and send a message. A reply means the setup works.
  3. The call and its cost appear under Usage Logs in NoviaHub.

Switch models

Section titled “Switch models”

Switch in the model selector in the chat window. You can pick from the models you added to the list in step 4.

Model IDs available on NoviaHub are listed under Models & Pricing, for example deepseek-v4-flash, claude-sonnet-5 and gemini-3-flash (all live on 2026-09-28). A model ID must match exactly.

Notes

Section titled “Notes”
  • If Check fails, look for a leftover old address in the other protocols’ address fields (such as localhost:3000 kept from a beta version), clear it and check again.
  • The API Key field takes several keys separated by commas; Cherry Studio rotates between them.
  • The one-click import doesn’t specify a protocol type, so Cherry Studio sets the provider’s default chat protocol to OpenAI Chat.

References

Section titled “References”

Checked on 2026-09-28: