Major improvements:
- Automatic GPU to CPU fallback with VRAM management
- Free PyTorch CUDA cache before each load attempt
- CPU-pure mode: op_offload=False, flash_attn=disabled, offload_kqv=False
(prevents llama.cpp from allocating ~606 MiB CUDA compute buffers)
- Progressive context reduction (4096 -> 2048 -> 512)
- Vision-Language model detection and filtering (Qwen3VL, LLaVA, etc.)
- Proactive CUDA DLL registration for Windows
- Always-verbose mode for native llama.cpp diagnostics
Ezequiel Tojo