perspectiva aiqbenchProduse AI

Liquid AI has introduced LFM2.5-24B-A2B, an open-weight Mixture-of-Experts model designed specifically to enable high-performance reasoning on local hardware like the Apple M5 Max. By activating only 2 billion parameters out of a 24-billion parameter total, this architecture aims to balance advanced agent capabilities with the strict latency and memory constraints of edge devices. This release represents a significant step forward in making sophisticated AI agents viable for on-device deployment without relying solely on cloud connectivity.

Sursă care acoperăMarkTechPost

Ce este aceastaLectura unui model asupra articolului și a acoperirii conexe, scrisă pentru aiqbench. Nu sunt cuvintele publicației.

Liquid AI Releases LFM2.5-24B-A2B: A 2B-Active MoE for On-Device Agents

MarkTechPost14 octombrie 20261 min de cititArticolul și perspectiva sunt în engleză.

Liquid AI released LFM2.5-24B-A2B, an open-weight 24B Mixture-of-Experts reasoning model with 2B active parameters. It targets local agents, scoring 48.09 on LiveClaw and decoding 184 tokens/s on an Apple M5 Max. The post Liquid AI Releases LFM2.5-24B-A2B: A 2B-Active MoE for On-Device Agents appeared first on MarkTechPost.

Citește articolul complet pe MarkTechPost

Citește pe MarkTechPost ↗Întreabă despre acest articol