Skip to main content
To use LaoZhang API in Cherry Studio, add an OpenAI-type provider, set the API address to https://api.laozhang.ai, then enter your LaoZhang API key and add models.

Add LaoZhang API

1

Create the provider

Open Cherry Studio, click Settings (the gear icon at the bottom left), go to Model Provider, and click Add at the bottom of the list. Name it “LaoZhang API” and choose OpenAI as the provider type.
2

Enter the key and address

  • API Key: your LaoZhang API key
  • API Address: https://api.laozhang.ai
Below the address, Cherry Studio previews the full request URL. It should read https://api.laozhang.ai/v1/chat/completions. An address ending in / stops Cherry Studio from adding /v1, and one ending in # is used exactly as typed; you normally need neither.
3

Check the connection

Click Check next to the API key and pick a model to test. When it passes, turn on the switch at the top right of the provider.
4

Add models

Click Manage to pick from the model list, or Add to enter a model ID such as gpt-6-sol. Model IDs must match the model catalog exactly.
Once added, switch to a LaoZhang API model from the model picker at the top of a chat.

Choose a model

Chats with MCP tools turned on can use any model in the table, including gpt-6-sol (since November 4, 2026).

FAQ

GPT-6 returns “Unsupported parameter: ‘max_tokens’”

Before November 4, 2026, GPT-6 models rejected max_tokens. LaoZhang API now accepts it, so you can keep the max-tokens setting. If the error still appears, turn off the max-tokens setting in the assistant or chat settings and try again.

With MCP tools on, GPT-6 returns “Function tools with reasoning_effort are not supported”

Before November 4, 2026, OpenAI’s rules let GPT-6 models call tools in Chat Completions only with reasoning effort set to none, so turning on MCP tools caused this error. LaoZhang API now handles these requests, and chats with MCP tools can use gpt-6-sol or gpt-6.1-sol directly. If you still see the error, check that the API address is https://api.laozhang.ai, then send support the time of the error and the model.

The connection check fails or returns 404

Make sure the API address is https://api.laozhang.ai and the preview shows only one /v1. For a 401, copy the key again and confirm the token is enabled in Tokens. For more, see Invalid API key or 404.