lfm25
Here are 5 public repositories matching this topic...
LFM2-powered research agent | Automated Related Work & citations | Semantic Scholar + arXiv
-
Updated
Jan 9, 2026 - Python
LFM2.5 1.2 billion reasoning language model runs on web browsers using Transformers.js and ONNX Runtime Web.
-
Updated
Jul 12, 2026 - JavaScript
LFM2.5 1.2 billion Japanese language model runs on web browsers using Transformers.js and ONNX Runtime Web.
-
Updated
Jul 12, 2026 - JavaScript
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.
-
Updated
Jun 7, 2026 - Cuda
Improve this page
Add a description, image, and links to the lfm25 topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the lfm25 topic, visit your repo's landing page and select "manage topics."