https://github.com/gurveervirk/blenderhelperbot-with-tidb
Streamlit app for a Blender Helper Bot using a pre-built TiDB vector store
https://github.com/gurveervirk/blenderhelperbot-with-tidb
blender blender-python bpy huggingfaceapi llama-index rag streamlit tidb tidbcloud vector-index
Last synced: 4 months ago
JSON representation
Streamlit app for a Blender Helper Bot using a pre-built TiDB vector store
- Host: GitHub
- URL: https://github.com/gurveervirk/blenderhelperbot-with-tidb
- Owner: gurveervirk
- License: apache-2.0
- Created: 2024-08-23T08:23:22.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-08-23T18:28:44.000Z (10 months ago)
- Last Synced: 2025-01-31T03:27:11.682Z (4 months ago)
- Topics: blender, blender-python, bpy, huggingfaceapi, llama-index, rag, streamlit, tidb, tidbcloud, vector-index
- Language: Python
- Homepage: https://blender-chatbot.streamlit.app/
- Size: 10.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# BlenderHelperBot
This is an _online RAG_ app, built using **TiDB vector store**, **llama-index** and **HuggingFace Inference API**, to help users with the **bpy** python package for **Blender** scripting.
TiDB vector store being used has already been fed **Blender 4.1 API docs**, for setting up the necessary context. The attached [`upload_to_tidb.py`](https://github.com/gurveervirk/BlenderHelperBot-with-TiDB/blob/main/upload_to_tidb.py) has been used for this purpose.
## Requirements
1) Get a free [HuggingFace API read token](https://huggingface.co/settings/tokens).
2) Get access to [this Mistral model](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3). This is being used as the underlying, instructed model.
## Known Issues
- Model may respond with some extra, unnecessary tokens / words, possibly due to it being a HuggingFace Inference API hosted model.