https://github.com/knuckles-team/genius-chatbot
Chatbot that uses any desired hugging face model or allows for scalable intelligence based on hardware limitations
https://github.com/knuckles-team/genius-chatbot
ai chatbot huggingface models python transformers
Last synced: about 1 year ago
JSON representation
Chatbot that uses any desired hugging face model or allows for scalable intelligence based on hardware limitations
- Host: GitHub
- URL: https://github.com/knuckles-team/genius-chatbot
- Owner: Knuckles-Team
- License: mit
- Created: 2022-12-23T03:19:58.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-05-09T03:48:40.000Z (about 2 years ago)
- Last Synced: 2025-04-05T06:46:04.869Z (about 1 year ago)
- Topics: ai, chatbot, huggingface, models, python, transformers
- Language: Python
- Homepage:
- Size: 122 KB
- Stars: 1
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Genius Chatbot
*Version: 1.10.4*
Chatbot that uses any hugging face model or OpenAI endpoint.
Local vector store supported with ChromaDB, or connect to a PGVector database
Allows for scalable intelligence tailored for hardware limitations
This repository is actively maintained - Contributions are welcome!
Contribution Opportunities:
- Support more vector databases
- Get list of models to choose from
- Output everything as JSON when selected
Usage:
| Short Flag | Long Flag | Description |
|------------|----------------------|---------------------------------------------------------------------------------|
| -h | --help | See Usage |
| -a | --assimilate | Assimilate knowledge from media provided in directory |
| | --batch-token | Number of tokens per batch |
| | --chromadb-directory | Number of chunks to use |
| | --chunks | Number of chunks to use |
| -e | --embeddings-model | [Embeddings model](https://www.sbert.net/docs/pretrained_models.html) to use |
| | --hide-source | Hide source of answer |
| -j | --json | Export to JSON |
| | --openai-token | OpenAI Token |
| | --openai-api | OpenAI API Url |
| | --pgvector-user | PGVector user |
| | --pgvector-password | PGVector password |
| | --pgvector-host | PGVector host |
| | --pgvector-port | PGVector port |
| | --pgvector-database | PGVector database |
| | --pgvector-driver | PGVector driver |
| -p | --prompt | Prompt for chatbot |
| | --mute-stream | Mute stream of generation |
| -m | --model | Copy [GPT4All](https://gpt4all.io/index.html) .bin file from the Model Explorer |
| | --max-token-limit | Maximum token to generate |
| | --model-directory | Directory to store models locally |
| | --model-engine | GPT4All LlamaCPP, or OpenAI |
Example:
```bash
genius-chatbot --assimilate "/directory/of/documents"
```
```bash
genius-chatbot --prompt "What is the 10th digit of Pi?"
```
```bash
genius-chatbot --prompt "Chatbots are cool because they" \
--model "wizardlm-13b-v1.1-superhot-8k.ggmlv3.q4_0.bin" \
--model-engine "GPT4All" \
--assimilate "/directory/of/documents" \
--json
```
Installation Instructions:
Install Python Package
Windows Prerequisites:
Visual Studio Code 2022
```bash
winget install -e --id Kitware.CMake
```
Ubuntu Prerequisites:
```bash
apt install -y pandoc
```
```bash
python -m pip install genius-chatbot
```
## Geniusbot Application
Use with a GUI through Geniusbot
Visit our [GitHub](https://github.com/Knuckles-Team/geniusbot) for more information
Installation Instructions with Geniusbot:
Install Python Package
```bash
python -m pip install geniusbot
```
Repository Owners:
