Closed sinedied closed 2 months ago
We have reproduced and this typically happens when the LLM returns a phone number.
This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.
I was also facing this issue, getting a response splitted in bullets. While trying to solve the issue, I found out that disabling the "Stream chat completion responses" option somehow solves the issue. Didn't investigate more, but the issue might be related to generating the array that contains the response.
I hope it can help somehow
I was also facing this issue, getting a response splitted in bullets. While trying to solve the issue, I found out that disabling the "Stream chat completion responses" option somehow solves the issue. Didn't investigate more, but the issue might be related to generating the array that contains the response.
I hope it can help somehow
Thank you for the context. We're evaluating moving that responsibility to the backend. Earlier when we delivered this sample, that was not possible, but we always planned to migrate the parser for performance reasons. I will reprioritize.
any update on the issue?
This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.
This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.
This issue was closed because it has been stalled for 7 days with no activity.
Sometime the parser fails at parsing some answers, for example for the question "How to contact a representative?" I get this:
This also occurs quite often when trying to cite sources when the backend is switched to the Python sample:
This issue is for a: (mark with an
x
)