Find answers from the community

Updated 7 months ago

Hey everyone,

At a glance
Hey everyone,

I've been using the llama-parse API to extract data from PDFs, and until last week, everything was working perfectly. However, this week, the results have become mostly garbled, with about 90% of the output looking like this:

:->|>11_NaN<|<-::->|>95_NaN<|<-::->|>93_NaN<|<-::->|>109_NaN<|<-::->|>106_NaN<|<-::->|>60_NaN<|<-::->|>109_NaN<|<-: :->|>81_NaN<|<-::

And only about 10% is coming through correctly. The code and document I'm using are exactly the same as before. Has anyone encountered a similar issue recently or have any suggestions on how to troubleshoot this?

Thanks in advance for your help!
1
L
p
S
10 comments
Hmm @pld or @Binary Brain might be able to help. Are you able to share a PDF causing the issue? What settings are you using?
Hey, sorry but unfortunately I can't share the PDF as it contains sensitive information. I'm using the basic configurations available here: https://github.com/run-llama/llama_parse

Thanks for reaching out!
Can you re-run the job and send us the JobID? I'd like to look at the logs.
Hey @Binary Brain, sure thing! I've re-run the job, and the job ID is '297983e8-623c-4174-868c-8aca0cb23a6d'. Thanks.
@pld This looks like some Unicode Markers from your code. Do you think you can help/fix?
Any news, @Binary Brain?
Hi! we will release a fix later today, hopefully it fix your issue.
@pld thanks for the update!
Hi! still working on the fix, its harder than planed, should land today hopefully.
Hey! Thank you for the update, looking forward to it.
Add a reply
Sign up and join the conversation on Discord