---
format: "aidr-story-markdown/v1"
id: "a6d60e2eeb246afd1e86dc9602678e1ae686d07bdd912e1fc498a6239961221a"
canonical_url: "https://aidr.today/a6d60e2e?lang=en"
title: "Meet Kyutai REKI: The 2.4B VLA Model Bringing Efficient Memory to Robot Manipulation"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-17T20:22:11.000Z"
category: "Models"
topics: ["agent"]
source_urls: ["https://www.marktechpost.com/2026/10/17/meet-kyutai-reki-the-2-4b-vla-model-bringing-efficient-memory-to-robot-manipulation/"]
summary: "Kyutai's REKI is a 2.4B vision-language-action model trained on public data that scores 90.4 on LIBERO-Plus and includes built-in memory for robot manipulation."
---

# Meet Kyutai REKI: The 2\.4B VLA Model Bringing Efficient Memory to Robot Manipulation

> [Open the canonical story](<https://aidr.today/a6d60e2e?lang=en>)

**Published:** 2026-10-17T20:22:11.000Z
**Category:** Models
**Topics:** agent

## Summary

Kyutai's REKI is a 2\.4B vision\-language\-action model trained on public data that scores 90\.4 on LIBERO\-Plus and includes built\-in memory for robot manipulation\.

## Sources

- [Story source](<https://www.marktechpost.com/2026/10/17/meet-kyutai-reki-the-2-4b-vla-model-bringing-efficient-memory-to-robot-manipulation/>)