Text generation Liquid AI

Lfm2.5 1.2B Thinking

Liquid LFM2.5 1.2B Thinking - hybrid conv+attention reasoning LLM optimized for edge deployment (fast CPU inference, low memory)

lfm2.5-1.2b-thinking

Make it part of your app.

One model. Your language.

use xybrid::ModelLoader;
use xybrid::ir::{Envelope, EnvelopeKind};

// LLMs need the llama.cpp backend: features = ["llm-llamacpp"]
let model = ModelLoader::from_registry("lfm2.5-1.2b-thinking").load()?;

// Run text generation
let input = Envelope::new(EnvelopeKind::Text("Explain quantum computing in simple terms.".into()));
let result = model.run(&input, None)?;

println!("{}", result.text().unwrap_or(""));

Built with Xybrid Quickstart

Available variants

Variant	Format	Quantization	Size
universalDEFAULT	GGUF	q4_k_m	697.0 MB