Update src/backend/model_loader.py

Update src/backend/model_loader.py

Update src/backend/model_loader.py

Update src/backend/model_loader.py

Update src/backend/model_loader.py

chore: increase max_new_tokens parameter to 4096 across chat model and inference functions for improved response generation

refactor: reorganize tokenizer input handling in Streamlit app and update model loading logic to default tokenizer revision to model revision when not specified

feat: add environment configuration, .gitignore, and project metadata; enhance model loading with optional parameters for model and tokenizer revisions

feat: add support for special model configurations in the model loading process

Implement initial project structure with Dockerfile, requirements, and Streamlit app. Added model loading and inference utilities, along with chat management features. Updated entry point and added new dependencies.