DeepYardDeepYard
L

LiquidAI LFM2.5

8B parameter edge-optimized language model with MoE architecture for on-device deployment

commercialFree

About

LFM2.5 is Liquid AI's 8-billion parameter language foundation model featuring Mixture-of-Experts (MoE) architecture optimized for edge deployment and on-device inference. Supports conversational AI across 9 languages with production-ready performance suitable for resource-constrained environments. With 500+ GitHub stars, it's designed for developers building autonomous agents that need to run locally without cloud dependencies.

Details

Typecoding-agent
Deploymentself-hosted
Supported Models

Tags

autonomousmulti-agentself-hostedopen-sourcepython