Local LLM, image&video&music generator, vibecode like cursor with local models on your phone
-
Updated
Aug 24, 2026 - C++
Local LLM, image&video&music generator, vibecode like cursor with local models on your phone
LFM2-powered research agent | Automated Related Work & citations | Semantic Scholar + arXiv
LFM2.5 1.2 billion reasoning language model runs on web browsers using Transformers.js and ONNX Runtime Web.
LFM2.5 1.2 billion Japanese language model runs on web browsers using Transformers.js and ONNX Runtime Web.
A minimal implementation of Liquid Foundation Model LFM 2.5 2.6B model
Building a Multimodal Model with LFM2.5 and Kimi-K2.6
A self-contained CUDA inference engine for LiquidAI/LFM2.5-8B-A1B (hybrid conv + GQA-attention MoE, 8.5B params, 1B active) targeting a single RTX 3060 (12 GB). No Python, no frameworks at runtime: a single .cu engine + a header-only byte-level BPE tokenizer.
To associate your repository with the lfm25 topic, visit your repo's landing page and select "manage topics."