Regarding the issue that the AI - chat request works fine, but the console reports an error

[miaomiao620]

Hello. I created a proxy named "grammar - chat - agent" by myself, and this proxy points to the "sprout - language - model" that I also created.
After I enter a request in the input box of the AI - chat - UI, the chat - agent can successfully call my request message to the sprout - language - model.
Moreover, the sprout - language - model can return a response normally, and the response can be displayed correctly in the input box of the AI - chat - UI. Everything seems to be working properly.

However, my console always reports an error. I'm not sure how to troubleshoot this error. It seems that there is a hidden request that has not been accepted by the RPC server.
The screenshot is as follows:

[msujew]

cc @sdirix

[sdirix]

From your post and stack trace it's not immediately clear where this issue is coming from. Can you do more debugging to check what is going wrong or provide a reproducible example so it's easier to investigate from our side?

It might be worth to add a simple "mock/stub" language model to the api-samples anyway (for example simply repeating back to the user what they wrote in their request). If you like you can contribute that and we can then discuss the error in the PR if it's reproducible that way.

[miaomiao620]

Thank you for your reply. I have found out the cause of the problem. It was an error caused by the Monaco Editor attempting to call a certain language service when the content changed. After I disabled the relevant property, there is no error message anymore.