---
title: microsoft/Florence-2-large
description: This is a continued pretrained version of Florence-2-large model with 4k context length, only 0.1B samples are used for continue pretraining. Florence-2, 777M parameters.
canonical_url: https://superlinked.com/models/microsoft-florence-2-large
last_updated: 2026-08-24
---

# microsoft/Florence-2-large

This is a continued pretrained version of Florence-2-large model with 4k context length, only 0.1B samples are used for continue pretraining, thus it might not be trained well. In addition, OCR task has been updated with line separator ('\n'). COCO OD AP 39.8

Source: [microsoft/Florence-2-large on HuggingFace](https://huggingface.co/microsoft/Florence-2-large)

## Overview

| Field | Value |
|-------|-------|
| Architecture | Florence-2 |
| Parameters | 777M |
| Tasks | Extract |
| Outputs | text_regions |
| License | mit |
| Inputs | image |

## Benchmarks

### COCO

Domain: general · Task: detection · Language: en

Object detection on COCO natural images

Corpus: 5,000 · Queries: 5,000

**Quality:** ap: 0.2940 · ap50: 0.4216 · ap75: 0.3160 · ar 100: 0.4918

[Reference](https://cocodataset.org/)