Using LFM2.5-VL-3B's Vision Capabilities (Liquid AI Guide)

Melvin Vivas · X post · 2026-08-14 · Open on X

Topics: LLM Fundamentals, AI Agents, Tool Use & MCP · Level: intermediate

Summary

Points to Liquid AI's guide for its small vision-language model, LFM2.5-VL-3B. The guide covers single- and multi-image prompting, OCR, document layout annotation, object detection and grounding, and tool calling.

Key points

• LFM2.5-VL-3B is a small 3B vision-language model from Liquid AI.
• Supported uses: single-image prompts and multi-image prompts.
• OCR and document layout annotation.
• Object detection and grounding (locating objects in an image).
• Tool calling from a vision model.

Resources mentioned

• Liquid AI Docs – Vision Capabilities guide · docs · docs.liquid.ai · free
  Guide to using LFM2.5-VL-3B for image prompts, OCR, layout annotation, grounding and tool calling.
• LFM2.5-VL-3B · tool · huggingface.co · free
  Liquid AI's lightweight vision-language model for screen and document understanding, grounding and tool calling.
  Also in: OCR Testing a Small Vision Model with LLM-Made Ground Truth (Melvin Vivas on X · notes), AIBackends 0.4.0: Liquid AI LFM2.5 Models on llama.cpp and Transformers (Melvin Vivas on X · notes), AIBackends Adds Support for LFM2.5-VL-3B (Melvin Vivas on X · notes), Quick test of Liquid AI's LFM2.5-VL-3B vision model (Melvin Vivas on X · notes) and 1 more

Try this

• Work through the guide's examples: OCR, multi-image prompts, grounding and tool calling.
• Build a document OCR and layout-extraction pipeline with a small local vision model.