UnicodeEncodeError: 'gbk' codec can't encode character '\u2081' in position 7318: illegal multibyte sequence

[ouening]

Failed in processing Chinese paper.

[igorlima]

It seems like the issue you're facing might be beyond the scope of py-zerox. However, providing more context and even a code snippet can be very helpful for others who might want to reproduce the issue. Below are a few issues that initially appeared to be beyond a library's scope:

• supporting Gemini #74 (comment)
• [Bug]: New llama 3.2 vision models from groq not working: prompting with images is incompatible with system messages BerriAI/litellm#6569 (comment)
• Add usage considerations to the docs #121 (comment)
• Handle ratelimit and retry #87 (comment)

Even so, a question like that can serve as documentation for anyone who might have a similar issue later. If you could share a PDF that triggers the UnicodeEncodeError and specify which model you're using, it would help identify the problem. Thanks!