Open harikt opened 2 months ago
Hi @harikt , it may related to modle that you used for this example. Could you try to get a response to the llamafile using this query directly to its endpoint (Postman might help)? I suggest you try both /completions
and /v1/chat/completions
. Let's see if the issue still presist.
@hfahrudin querying directly doesnot have any issues.
If you have no issue on both endpoints, then the problem may lie somewhere else. Could you reproduce the issue consistently? please set the seed number so we could try to reproduce locally and consistently.
Checked other resources
Example Code
I have the following code and the llamafile is running on default port 8080.
The poetry file looks like below
When executing this is calling multiple questions and not ending,.
This was noticed only after using the stream. I believe this is a bug.
Error Message and Stack Trace (if applicable)
Recursively calling the queries that are not asked.
Description
No errors thrown. It waits and continues asking question and showing answers.
System Info
"pip freeze | grep langchain"
langchain==0.2.16 langchain-community==0.2.16 langchain-core==0.2.38 langchain-text-splitters==0.2.4
python -m langchain_core.sys_info
System Information
Package Information
Optional packages not installed
Other Dependencies