Enhance Text Analyzer Prompt for Maximum Intelligence and Flexibility

Fix syntax error in verify_video endpoint and ensure challenge endpoint exists

Final fix: Use gemini-flash-latest for Chatbot (Proven quota available)

Update Chatbot to gemini-2.0-flash-exp with safety settings and robust handling

Revert to models/gemini-flash-latest to fix quota issues

Update all models to models/gemini-2.0-flash based on debug API response

Force usage of gemini-1.5-flash for stability and update dependencies

Update google-generativeai version and add debug endpoint

Revert to gemini-1.5-pro and improve fallback logic for short text

Use gemini-1.5-flash-latest (stable and confirmed working)

Fix model name to gemini-1.5-pro-latest (correct format)

Fix model format with models/ prefix for all Gemini models

Upgrade all analyzers to gemini-1.5-pro for maximum intelligence

Fix all models to use valid gemini-1.5-flash model name

Revert to gemini-flash-latest (smarter) for analyzers, keep 1.5-flash for chatbot

Standardize all models to gemini-1.5-flash for 85-90% accuracy and stable quota

Use gemini-1.5-flash for chatbot - better quota and still smart

Add /chat endpoint for chatbot

Use stable Bookworm base image

Fix: update deprecated package libgl1-mesa-glx to libgl1

Fix README encoding

Add Hugging Face Spaces configuration

Deploy Factify Models w/ Docker support

initial commit