| streamlit | |
| transformers>=4.38.0 # Or even try without a minimum version to get the absolute latest | |
| torch>=2.0.0 | |
| accelerate>=0.20.0 | |
| sentencepiece | |
| bitsandbytes # Good to have if you later want to try 8-bit/4-bit quantization for larger models | |
| streamlit | |
| transformers>=4.38.0 # Or even try without a minimum version to get the absolute latest | |
| torch>=2.0.0 | |
| accelerate>=0.20.0 | |
| sentencepiece | |
| bitsandbytes # Good to have if you later want to try 8-bit/4-bit quantization for larger models | |