Replace Novita AI with ZeroGPU Chat API (RunPod)

Fix indentation issues in prepare_context_for_llm and _truncate_to_tokens methods

FINAL FIX: Correct except clause indentation - ensure proper alignment with try statement

Fix return statement and else clause indentation in response handling

Fix else clause indentation at line 153 - align with if stream

Fix else clause indentation in response handling

Fix raise statement indentation in exception handler

CRITICAL FIX: Correct except clause indentation at line 58 - align with try statement

Add logging for context window configuration and improve max_tokens calculation debugging

Fix IndentationError at line 378: correct indentation in _truncate_to_tokens method

CRITICAL FIX: Correct except clause indentation at line 57 - align with try statement

Fix raise statement indentation

Fix IndentationError: correct else clause alignment - ensure else aligns with if stream

Fix IndentationError: correct else clause alignment at line 152

Fix critical syntax error: correct except clause indentation at line 57

Clean up try-except block formatting

Fix encoding issue: replace special character in logger message

Fix except clause indentation in llm_router.py

Fix syntax errors in llm_router.py - Fix indentation issues in try/except blocks - Fix else clause alignment

Fix max_tokens calculation to respect model context window (5000 tokens) - Add dynamic max_tokens calculation based on input size - Add novita_model_context_window configuration - Prevents 400 errors when input tokens exceed available output space

Fix max_tokens calculation to respect model context window

Integrate Novita AI as exclusive inference provider - Add Novita AI API integration with DeepSeek-R1-Distill-Qwen-7B model - Remove all local model dependencies - Optimize token allocation for user inputs and context - Add Anaconda environment setup files - Add comprehensive test scripts and documentation

Fix: DynamicCache compatibility, dependencies, and Docker configuration

Fix: BitsAndBytes compatibility and error handling

Phase 1: Remove HF API inference - Local models only

Fix infinite fallback loop in local model loading

Update model IDs to use Cerebras deployment and add gated repository error handling

Security Enhancements: Production WSGI, Rate Limiting, Security Headers, Secure Logging

Initial commit: Research AI Assistant API