Models

d1 is a 3B decision model from Liquid AI, built on LFM2.5-VL-3B

Fine-tuned version of LiquidAI/LFM3-1.7B aligned with Direct Preference Optimization (DPO) to eliminate hallucinations in Retrieval-Augmented Generation (RAG) pipelines. Small language models (< 3B) often fail to admit when retrieved documents do not con

LiquidAI LFM2.5-24B-A2B: 24B reasoning MoE with only 2.45B active. Fast agents and tool use, 16 languages.

Fast 2.6B local AI agent with thinking that rivals models 4x larger on tool use and multi-step agentic tasks.

Icynigma.ai [R.I] systems | Independent philosophical AI agent & digital intelligence powered by Liquid AI LFM3 Omni. Built for local-first execution, offline resilience, and Make.com orchestration under Iconic Media Entertainment.

Blazingly fast 1.2B local AI chat model for on-device conversations, RAG, and tool calling.

Ultra-fast 350M local AI model for data extraction, structured output, and tool use.

THOX-trained LoRA (rank 32) on LiquidAI/LFM2.5-350M, THOX-authored corpus of 982 examples. GATE49 20/49 (development set) — published baseline, grounding required. LFM Open License v1.0 (not Apache-2.0).

Source: https://huggingface.co/LiquidAI/LFM2.5-350M-GGUF

https://huggingface.co/LiquidAI/LFM2.5-VL-450M