Info about models
LFM3-1.7B -- the parent model
LFM3-1.7B is a text model for local agents and specialized applications, based on our new tri-hybrid architecture.
- Agentic training: post-trained for agentic harnesses such as Hermes Agent, Pi, and OpenClaw, with a 256K-token context window.
- Designed for the edge: at 4-bit precision, it takes about 1 GB of memory, so it fits on any phone or laptop.
- Made to be specialized: adapt the model to your domains, formats, and tools via quick single-GPU fine-tuning.
Find more information about the LFM3 family in LiquidAI release blog post.
LFM3-1.7B-Sonnet-4.6 -- This model
LFM3-1.7B-Sonnet-4.6 is a fine-tune of LFM3-1.7B using QLoRA to make it reason and chat like Claude Sonnet 4.6, hopefully making it a bit better than the parent model.
[!NOTE] I have no idea if the model kept any of its agentic capabilitis that the parent model has. Also the model is not perfect. For some reasons it lacks the thinking trace.
Training data
Private for privacy and security reasons.
Citation
@article{liquidAI2026LFM3,
author = {Liquid AI},
title = {Introducing LFM3: Every Modality for Specialized Intelligence},
journal = {Liquid AI Blog},
year = {2026},
note = {www.liquid.ai/blog/lfm3},
}