---
title: When LLMs communicate without intermediate text
author: Martin Ebner
date: 2026-10-27
url: https://postgenerative.at/indications/2026-10-27/when-llms-communicate-without-intermediate-text
lang: en
type: Indication
summary: "Cache-to-Cache introduces a new way for LLMs to communicate: models transfer internal representations directly through their KV caches, instead of first writing messages to each other."
tags:
  - Communication
  - Research
  - Software
  - Collaboration
sources:
  - title: "Tianyu Fu et al.: Cache-to-Cache: Direct Semantic Communication Between Large Language Models. ICLR 2026; arXiv:2510.03215, Version 2 vom 02.03.2026."
    url: https://arxiv.org/abs/2510.03215
  - title: Volltext der analysierten Fassung (PDF, Version 2).
    url: https://arxiv.org/pdf/2510.03215v2
license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
alternate:
  de: https://postgenerativ.at/indizien/2026-10-27/wenn-llms-ohne-zwischentexte-kommunizieren
---

# When LLMs communicate without intermediate text

Tianyu Fu and co-authors introduce Cache-to-Cache (C2C), a direct exchange of internal context representations between language models. A trained module translates and combines their KV caches – stored states used in attention computation. This removes the need to generate intermediate messages token by token. [1]

The study reports higher accuracy and an average 2.5-fold speedup over text communication. The approach requires access to model internals; incorrect information can propagate. Scaling to many models remains unresolved. [2]

Our interpretation: this example shifts attention from individual chatbots to cooperating models. Their collaboration could increasingly operate through technical representations.

[Read the study on arXiv](https://arxiv.org/abs/2510.03215)