Cross-Model KV Cache Transfer: Skip Re-Prefill When Switching LLMs

The Problem Nobody Had Solved Every production LLM system eventually confronts the same wall: you want to swap between models — route a cheap query to the 14B, an expensive one to the 32B — but the mo