tejasgaikwad16092002/LLM-Reasoning-Router

Add character-by-character typing effect for ChatGPT-like streaming

Restore SSE streaming for real-time chat responses

Fetch fresh metrics after each request for instant stats update

Fix model detection for Pro model stats and cost calculation

Use non-streaming endpoint for reliable metrics logging

Fix imports and add error traceback for debugging

Log metrics before streaming done message

Fix streaming metrics with dedicated database session

Fix metrics persistence for streaming endpoint

Add deployment configurations and test suite

Create interactive frontend with real-time streaming UI

Implement quality validation and escalation system

Build REST API layer with OpenAI-compatible endpoints

Add routing engine and Gemini LLM client integration

Implement prompt complexity analyzer with signal detection

Add database layer with SQLAlchemy async support

Initialize FastAPI project structure