← Help Center· Desktop graphics client

Connect Newcclm with Cherry Studio

Cherry Studio is a desktop chat client with a Chinese interface. It supports custom OpenAI compatible service providers, and you can use our gear by adding one.

Important: Do not change the built-in OpenAI provider. Create a new custom provider, otherwise Cherry Studio will request according to the official OpenAI rules and cannot connect to us.

1. Add a custom provider

OpenSet up → model services, find the bottom of the listTambah, Fill in the nameNewcclm, provider type selectionOpenAI(it is just a protocol format and does not represent the manufacturer).

Fill in these values

API URLhttps://sccclm. com/v1

API keyYour sk- * key

Model Namenewcclm-fast

Individual versions of Cherry Studio auto-replenish the path. If you fill inhttps://sccclm.com/v1hindcast 404, Just try to fill inhttps://sccclm.comTry both, whichever works best for you.

2. Add a model

ClickAdd model, Fill in the model ID with our stall name. Add both, and cut them below when used:

Former ID	When to use
`newcclm`	日常主力，不确定选哪个就用它
`newcclm-fast`	Reply to emails, change copy, organize materials, ask daily questions
`newcclm-coding`	Complex analysis, reading long documents, writing code
`newcclm-free`	免费档位，试用

Don't fill ingpt-4o、claude-*、deepseek-*Such a name. We only recognize our own stall name, and filling in anything else will report that the model does not exist.

3. Don't match it like this

The following are the most common mistakes:

Do not change the built-in OpenAI/DeepSeek provider

The addresses of built-in entries are hard-coded. The address was changed but the "provider type" was not changed. The client may still make requests according to built-in rules. The correct approach isCreate a new provider。

Do not add/chat/completions after the address

# Wrong
https://sccclm.com/v1/chat/completions

# Correct
https://sccclm.com/v1

Do not overlap/v1

# Error - will become /v1/v1/...
https://sccclm.com/v1/v1

Do not include the key in system prompts or conversations

The key is only populated in the API Key field of the settings page.

4. Take a test

After the configuration is completed, send "Hello" in Cherry Studio. It'll be fine if you can come back. If an error is reported, first run this in the terminal to distinguish whether it is a key issue or a client setting issue:

curl https://sccclm.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
-d '{"model":"newcclm-fast","messages":[{"role":"user","content":"Hello"}]}'

Results	Explain where the problem is
This line can be understood, the client does not understand	Client setup issues, see sections 1 and 3
This one doesn't make sense either.	Key or network issue, not related to Cherry Studio

Still not working

Send us these three things:

1. Take a screenshot of the API address column in the settings page (key coding)
2. Complete error reported by Cherry Studio
3. The output of curl above

Back to Help Center · Chatbox Configuration →