I have read and followed all the instructions provided in the README.md.
I am on the latest version of both Open WebUI and Ollama.
I have included the browser console logs.
I have included the Docker container logs.
I have provided the exact steps to reproduce the bug in the "Steps to Reproduce" section below.
Expected Behavior:
Upload, read and store text documents (as .pdf, .docx, .txt, etc) to be read and used as a context for users chat prompt. Using Chroma VB collection as storage.
Actual Behavior:
The storage of bigger documents crashes the application.
Description
Bug Summary:
Small documents works fine. Bigger documents fail at collection.upsert() method and crashes the application. Every collection accepts a maximum of 98 items and the example pdf has 427 parts after the embedding process.
The first example shows the upsert method (when using the insert method it happens the same way) and the vscode log when the application crashes. The upload part works fine, the file is completely read, the collection is found (overwrite is True so the collection is erased and recreated) but the collection.upsert() dont save more than 98 items per collection. Is there any way to increase this limit?
Originally created by @LuizReina on GitHub (Nov 1, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/6638
## Installation Method
git clone
## Environment
- **Open WebUI Version:** 0.3.30
- **ChromaDB Version:** 0.5.5
- **Operating System:** Windows 10 Pro and Windows 11 Pro
- **Browser (if applicable):** Chrome 130.0.6723.92
- **Model used:** Claude 3.5 Sonnet
- **Embedding model:** sentence-transformers/all-MiniLM-L6-v2
**Confirmation:**
- [x] I have read and followed all the instructions provided in the README.md.
- [ ] I am on the latest version of both Open WebUI and Ollama.
- [ ] I have included the browser console logs.
- [ ] I have included the Docker container logs.
- [x] I have provided the exact steps to reproduce the bug in the "Steps to Reproduce" section below.
## Expected Behavior:
Upload, read and store text documents (as .pdf, .docx, .txt, etc) to be read and used as a context for users chat prompt. Using Chroma VB collection as storage.
## Actual Behavior:
The storage of bigger documents crashes the application.
## Description
**Bug Summary:**
Small documents works fine. Bigger documents fail at collection.upsert() method and crashes the application. Every collection accepts a maximum of 98 items and the example pdf has 427 parts after the embedding process.
## Reproduction Details
**Steps to Reproduce:**
- Upload the example pdf;
- Wait for uploading process;
- Check vscode logs.
## Logs and Screenshots
[07373a5b-a891-46e3-82dd-3be983a2775d_Nota_Fiscal.pdf](https://github.com/user-attachments/files/17603577/07373a5b-a891-46e3-82dd-3be983a2775d_Nota_Fiscal.pdf)
**Screenshots/Screen Recordings (if applicable):**
Example 1:
## Additional Information
The first example shows the upsert method (when using the insert method it happens the same way) and the vscode log when the application crashes. The upload part works fine, the file is completely read, the collection is found (overwrite is True so the collection is erased and recreated) but the collection.upsert() dont save more than 98 items per collection. Is there any way to increase this limit?
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @LuizReina on GitHub (Nov 1, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/6638
Installation Method
git clone
Environment
Open WebUI Version: 0.3.30
ChromaDB Version: 0.5.5
Operating System: Windows 10 Pro and Windows 11 Pro
Browser (if applicable): Chrome 130.0.6723.92
Model used: Claude 3.5 Sonnet
Embedding model: sentence-transformers/all-MiniLM-L6-v2
Confirmation:
Expected Behavior:
Upload, read and store text documents (as .pdf, .docx, .txt, etc) to be read and used as a context for users chat prompt. Using Chroma VB collection as storage.
Actual Behavior:
The storage of bigger documents crashes the application.
Description
Bug Summary:
Small documents works fine. Bigger documents fail at collection.upsert() method and crashes the application. Every collection accepts a maximum of 98 items and the example pdf has 427 parts after the embedding process.
Reproduction Details
Steps to Reproduce:
Logs and Screenshots
07373a5b-a891-46e3-82dd-3be983a2775d_Nota_Fiscal.pdf
Screenshots/Screen Recordings (if applicable):
Example 1:
Additional Information
The first example shows the upsert method (when using the insert method it happens the same way) and the vscode log when the application crashes. The upload part works fine, the file is completely read, the collection is found (overwrite is True so the collection is erased and recreated) but the collection.upsert() dont save more than 98 items per collection. Is there any way to increase this limit?