Layer-wise CPU-GPU Scheduling for LLM Inference on Memory-Limited Consumer GPUs

Connections are not prepared for this record yet, or its source metadata has changed. The original record remains available while background snapshots are rebuilt.