# Processing PDF's into Assistant starts throwing OCR error

**URL:** <https://community.pinecone.io/t/processing-pdfs-into-assistant-starts-throwing-ocr-error/8211>\
**Category:** Support\
**Tags:** pinecone-assistant\
**Created:** [July 7, 2025, 12:38pm UTC](https://community.pinecone.io/t/processing-pdfs-into-assistant-starts-throwing-ocr-error/8211 "2025-07-07T12:38:20Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![agentgill](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/agentgill/32/4386_2.png) [@agentgill](https://community.pinecone.io/u/agentgill)\
**Post date:** [July 7, 2025, 12:38pm UTC](https://community.pinecone.io/t/processing-pdfs-into-assistant-starts-throwing-ocr-error/8211/1 "2025-07-07T12:38:20Z")

</div>

We have a python script which loads files locally stored PDF’s into a pinecone assistant - nothing too clever .

After what appears to be a random number of API calls and PDFs uploaded successfully, it starts to throw this error. If I stop it and process the file individually, it’s fine. If I restart it, it’s obviously fine, but it’s just weird. I’m trying to load about 1000 PDFs and after about 50 or so, it starts to then throw this error, so it’s like some weird limit

```auto
def upload_to_pinecone(assistant, file_path: Path, metadata: dict) -> bool:
    """Upload a file to Pinecone Assistant with metadata."""
    try:
        response = assistant.upload_file(
            file_path=str(file_path), metadata=metadata, timeout=None
        )
        print(f"Uploaded successfully {response}")
        return True
    except Exception as e:
        print(f" Upload failed: {e}")
        return False

```

---

<div class="post-metadata">

**Author:** ![agentgill](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/agentgill/32/4386_2.png) [@agentgill](https://community.pinecone.io/u/agentgill)\
**Post date:** [July 7, 2025, 12:38pm UTC](https://community.pinecone.io/t/processing-pdfs-into-assistant-starts-throwing-ocr-error/8211/2 "2025-07-07T12:38:55Z")

</div>

All files are OCR’d BTW

---

<div class="post-metadata">

**Author:** ![arjun](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/arjun/32/2595_2.png) [@arjun](https://community.pinecone.io/u/arjun)\
**Post date:** [July 9, 2025, 6:36pm UTC](https://community.pinecone.io/t/processing-pdfs-into-assistant-starts-throwing-ocr-error/8211/3 "2025-07-09T18:36:29Z")

</div>

Hi @agentgill!

What’s the specific error you are getting, and how large are those first 50 files? Could be storage related on the starter tier, which you are capped to 1GB for.

> **[Pricing and limits - Pinecone Docs](https://docs.pinecone.io/guides/assistant/pricing-and-limits#usage-exceeds-monthly-minimum)**
>
> Search through billions of items for similar matches to any object, in milliseconds. It's the next generation of search, an API call away.

Let us know that specific error, and we can help further!

-Arjun