Local LLM, image&video&music generator
-
Updated
Nov 19, 2026 - C
Local LLM, image&video&music generator
A skill and CLI that lets small local models like LFM3-1.7B make 3D printable parts, from words or a picture, even through context compaction.
A dependency-free C inference engine for Liquid AI's LFM2 / LFM2.5 models. Terminal chat, an OpenAI-compatible server, a web agent and a terminal agent, all in one executable.
Run LiquidAI's LFM3 collection fully in your browser, on the CPU: chat, images, speech in and out, and d1's structured answers, at up to 97 tokens/s. Rust compiled to WebAssembly with fusor-ai. No server, no GPU.
A local LFM3 powered agent for everyday tasks, browser use and vision
LFM3-1.7B (pruned, 4-bit) Core ML packages for the Apple Neural Engine, with placement, speed and accuracy evidence. Early release.
Application Flutter de démonstration pour exécuter LFM3-1.7B (texte) et LFM3-Audio-2B (voix) entièrement en local, sans serveur
在安卓(骁龙 8 Gen 1)上跑 Liquid AI LFM3-1.7B:预编译 arm64 二进制、内置聊天页、实测性能与调优记录
Download Audio Report AI for iOS: local-only transcription and LFM3 report drafting with bundled offline models.
LFM3-1.7B QAD-Q4_0 × Serena MCP agent bridge on Jetson Orin Nano 8GB — one-shot HTML5 game generation with real tool calls, 256K context (q4_0 KV), ~31.7 tok/s, BFCL v2 92.3%; canonical staged-build method (stub skeleton → one stub per turn, tripled output).
A 1.1 GB LLM (LFM3 1.7B) takes a 30-task exam fully offline on a Mac mini and scores 29/30. Tasks, graders, raw outputs and results.
Fine-tune LFM3 into decision models with Unsloth for B2B lead qualification and reply routing, served to Claude Code over MCP. Companion to the tutorial on raoofaltaher.com.
Windows GUI for LFM3 Omni with Tool usage for local image creation and local wikipedia knowledge (DE / EN)
Liquid AI LFM3の全6モデルをColab T4で試す実験ノート。日英ガイド、実測VRAM・速度・失敗例、再現用CLIを公開。
Public install-and-run distribution for the local-first Audio Report AI Ubuntu desktop app
Local low-latency voice assistant on LFM3-Audio 2B and llama-server: incremental audio, turn-taking, and barge-in.
To associate your repository with the lfm3 topic, visit your repo's landing page and select "manage topics."