Nebius Token Factory
Mini toolsLFM2.5 1.2B Thinking
By Liquid AI
LFM2.5-1.2B-Thinking is the general purpose reasoning member of the LFM2.5 line, which uses a hybrid architecture tuned for on device deployment. The Thinking variant keeps the 1.2B parameter footprint but devotes training to multi step reasoning, tool use and RAG style workloads, offering up to 32K context and explicit “reasoning tokens” separate from final completions. On OpenRouter it is provided as a free tier endpoint, giving agents a small, edge ready model that still performs competitively with much larger LLMs.
New
Text
Gen 7
Overview
LFM2.5-1.2B-Thinking is Liquid AI’s 1.2B parameter reasoning-focused variant of its LFM2.5 edge family, optimized for agentic workflows, RAG and data extraction with 32K context, and exposed on OpenRouter as a free model.
About Liquid AI
Liquid AI is an MIT spin-off building efficient general-purpose AI models (Liquid Foundation Models, or LFMs) that run on edge devices with less memory and power.
They recently raised $250M in Series A funding to scale model development and deployment.
Website:
liquid.ai
Tools using LFM2.5 1.2B Thinking
No tools found for this model yet.
Last updated: February 3, 2026
