Speed up: eager attn + KV cache; drop chat retries to 1; remove MiniCPM-o voice UI artifact

Log + surface RuntimeErrors from witness chat too (still 503 for those)

Surface witness chat failures with traceback + error class in 500 detail

Surface zerogpu backend load_error / load detail in setup status

Eagerly import zerogpu_backend on Spaces so @spaces.GPU is registered before startup scan

Force demo.launch to bind 0.0.0.0:$PORT on HF Spaces (CLI hot-reload ignores GRADIO_SERVER_NAME)

Expose `demo` at module scope so Gradio SDK runner can launch the gr.Server app

Gate setup/llama subprocess paths behind provider check; allow zerogpu_transformers

Ship Phantom Grid Docker Space