LFM2.5-1.2B-Thinking
Sub-1GB reasoning variant of the LFM2.5 1.2B on-device model.
About
LFM2.5-1.2B-Thinking is the reasoning-oriented variant of Liquid AI's 1.2B LFM2.5 model, released on January 20, 2026, weeks after the base 1.2B launch. It brings explicit reasoning capabilities to the company's on-device line in a footprint under 1GB, targeting phones, robots, and home automation where cloud access is impractical.
Like the rest of the LFM2.5 family, it is built on Liquid AI's hybrid convolution-attention architecture and improved instruction following, tool use, and multilingual vision over the LFM2 generation. Weights are available on Hugging Face under Liquid AI's open-weight program; verify the exact weight license on the model card.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 106; Trending score 66/100 from researcher assessment.