Skip to content

Florence-2-large-ft

Prompt-driven vision and vision-language tasks

Visit official site

What Florence-2-large-ft is

The model is described as a Microsoft vision foundation model that uses text prompts for vision and vision-language tasks.

What you can do with it

  • Write a text description of an image
  • Locate objects inside an image
  • Split an image into regions or objects

How it works

  • transformers and PIL are installed first
  • An AutoProcessor loads prompts and images
  • Inputs are converted before the model runs

Pros & cons

✓ What we like

  • Simple text prompts drive many task types
  • Region proposals and region OCR are included

! What to watch out for

  • transformers and PIL are installed before use
  • A text prompt and an image are prepared together

FAQ

What tasks are listed?

The material lists captioning, object detection, segmentation, region proposals and OCR.

What data is cited?

The material cites 5.4 billion annotations over 126 million images in FLD-5B.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review