# Nvidia study maps KV caches between compatible LLMs

_Friday, August 21, 2026 at 12:33 PM EDT · AI, Science · Latest · Tier 2 — Notable_

![Nvidia study maps KV caches between compatible LLMs — Primary](https://images.ctfassets.net/jdtwqhzvc2n1/6VtMw5wL0u0P76DJndqWOa/6f16765a7e816dd34a8b7f9cfb425eb0/kv_cache_transfer.jpg?w=800&q=75)

Researchers at Nvidia reported a method for transferring an LLM's KV cache between compatible models, avoiding a fresh prefill when an agentic workflow switches model sizes. In tests across matched-KV model pairs, the linear mapper ran 2.7 to 25 times faster than re-prefilling and retained 73% to 98% of target-model accuracy for four of six pairs.

## Sources

- [VentureBeat](https://venturebeat.com/technology/nvidia-finds-that-simple-linear-math-can-replace-costly-ai-model-handoffs)

---
Canonical: https://techandbusiness.org/newswire/nuqhqCX-aRd_kOvrTfhFRT
Retrieved: 2026-08-21T19:07:21.399Z
Publisher: Tech & Business (techandbusiness.org)