Want to use Chinese-developed models such as Kimi or Qwen for their Chinese-language capabilities, faster responses, or lower costs—but can’t find a ready-made node in n8n?
What if there were a “universal key” that could connect most major Chinese models to n8n, without writing a single line of code?
That key is OpenAI API compatibility.
It might sound technical, but this guide breaks it down into simple steps you can follow with just a few clicks. Let’s give your n8n workflow access to more models!
- A quick look at the setup
Here are some Chinese model services configured in the example: ModelScope, Kimi, and Tencent Hunyuan.
Next, we’ll use Kimi K2 as an example and walk through the configuration step by step.
-
Get an API key for Kimi K2 through Tokatlas
Visit Tokatlas and create an API key.
Then copy the API address:
https://api.tokatlas.com -
Configure n8n
As usual, let’s add our AI Agent node.
Use a Chat Trigger, then configure the AI Agent node.
Click Chat Model, scroll to the bottom, and select OpenAI Chat Model. Any model with an OpenAI-compatible API can be configured using this node.
Next, configure the credentials in n8n by entering your API key and Base URL.
For Kimi, enter the Base URL we noted earlier:https://api.tokatlas.com
Want to give it a try?
We’re offering 20 developers free API credits to try Tokatlas with their own projects. Tell us what you’re building—we’d love to hear your feedback!
Visit [https://app.tokatlas.com]or email [affiliates@tokatlas.com]to apply.