What are the hardware requirements for Florence 2 DocVQA?

Florence 2 DocVQA requires approximately 3.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Florence 2 DocVQA and how large is it?

Florence 2 DocVQA is developed by HuggingFaceM4, a model with 822.9m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Florence 2 DocVQA?

Florence 2 DocVQA is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Florence 2 DocVQA by HuggingFaceM4 — VRAM 3.3GB

Name: Florence 2 DocVQA
Author: HuggingFaceM4

Florence 2 DocVQA is an open-source language model by HuggingFaceM4. Features: 822.9m LLM, VRAM: 3.3GB, LLM Explorer Score: 0.14.

Arxiv:1910.09700 Autotrain compatible Base model:finetune:microsoft/... Base model:microsoft/florence-... Custom code Florence2 Image-text-to-text Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/HuggingFaceM4/Florence-2-DocVQA

Florence 2 DocVQA Parameters and Internals

Model Type

image-text-to-text

Supported Languages

English (High)

Training Details

Data Sources:

Docmatix

Data Volume:

5% of Docmatix data

Methodology:

Fine-tuning

Training Time:

1 day

LLM Name	Florence 2 DocVQA
Repository 🤗	https://huggingface.co/HuggingFaceM4/Florence-2-DocVQA
Base Model(s)	Florence 2 Base Ft microsoft/Florence-2-base-ft
Model Size	822.9m
Required VRAM	3.3 GB
Updated	2025-08-20
Maintainer	HuggingFaceM4
Model Type	florence2
Model Files	3.3 GB
Model Architecture	Florence2ForConditionalGeneration
Model Max Length	1024
Transformers Version	4.41.2
Tokenizer Class	BartTokenizer
Padding Token	<pad>
Vocabulary Size	51289
Torch Data Type	float32
Errors	replace

Rank the Florence 2 DocVQA capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Florence 2 DocVQA

Best Alternatives	Context / RAM	Downloads	Likes
Florence 2 Large Ft Bf16	0K / 1.6 GB	63	1
Florence Vl3	0K / 3.3 GB	3	2

Note: green Score (e.g. "73.2") means that the model is better than HuggingFaceM4/Florence-2-DocVQA.