Skip to main content

Image Captioning

Detailed descriptions, alt-text generation, and visual summarization.

OCR & Extraction

Text recognition, form parsing, and document digitization.

Visual Reasoning

Scene understanding, spatial relations, and visual Q&A.

Continuous Capture

Always-on activity recognition and scene monitoring on-device.

LFM3 Models   Latest release

LFM3-VL pairs the LFM3 language backbone with a new in-house vision encoder that combines convolution and attention. It brings video understanding to LFM vision models for the first time, and reasons step by step before answering.

LFM3-VL-2B

2B RecommendedVideo understanding, thinking, and spatial and embodied reasoning.

LFM2.5 Models

LFM2.5-VL models are refreshed versions of LFM2-VL, built on new LFM2.5 language backbones (LFM2.5-350M, LFM2.5-1.2B-Base, and LFM2.5-2.6B) with SigLIP2 NaFlex vision encoders. They improve instruction following and multilingual vision understanding, and support function calling.

LFM2.5-VL-3B

3B Most capableGrounding, screen and document understanding, and function calling.

LFM2.5-VL-1.6B

1.6B BalancedBalance of speed and quality for most use cases.

LFM2.5-VL-450M

450M FastestCompact vision model for edge deployment and fast inference.

Guides

Learn how to use these models in your application:

Vision Capabilities

Describe and compare images, read text, parse document layouts, point at objects, and detect objects.

Video examples

Describe videos and answer questions about them, from the frames alone or with the soundtrack.

Tool Calling

Call functions from text, images, and speech.

Examples

Explore practical implementations using vision models:

Image Understanding with Vision Language Models

Analyze images, answer visual questions, and generate descriptions using LFM2.5-VL-1.6B on Android with Jetpack Compose and Coil.Platform: Android · Uses: LFM2.5-VL-1.6B

Invoice Extractor Tool

Extract structured payment data from invoice PDFs using LFM2.5-VL-1.6B with file monitoring and 100% local processing.Platform: Desktop · Uses: LFM2.5-VL-1.6B

Real-Time Video Captioning

Generate video captions directly in-browser using LFM2.5-VL-1.6B with WebGPU acceleration and ONNX Runtime Web.Platform: Web · Uses: LFM2.5-VL-1.6B

Fine-tune for Car Maker Identification

Learn to fine-tune LFM2-VL models (450M, 1.6B, 3B) with LoRA, structured generation, and evaluation pipelines for image classification.Platform: Desktop · Uses: LFM2-VL-450M, 1.6B, 3B