EraX-VL-2B-V1.5 Open-source Multimodal Model - Supports Multilingual Images and Image/Text to Text Conversion

Erax VL 2B V1.5 I1 GGUF

Developed by mradermacher

EraX-VL-2B-V1.5 is a multimodal foundation model supporting Vietnamese, English, and Chinese, with capabilities for image-to-text and image-text-to-text conversion.

Image-to-Text Supports Multiple LanguagesOpen Source License:Apache-2.0 #Multimodal Image-to-Text #Vietnamese OCR #Insurance Document Processing

Downloads 467

Release Time : 12/29/2024

Model Overview

This is a multimodal vision-language model focusing on image-to-text and image-text-to-text conversion tasks, particularly suitable for fields like insurance and optical character recognition (OCR).

Model Features

Multilingual Support

Supports text processing in three languages: Vietnamese, English, and Chinese

Multimodal Capability

Can process both image and text inputs to achieve image-to-text conversion

Diverse Quantized Versions

Offers multiple quantized versions to accommodate different hardware and performance needs

Model Capabilities

Image-to-text

Image-text-to-text

Multilingual processing

Optical Character Recognition (OCR)

Use Cases

Insurance Industry

Insurance Document Processing

Automatically identifies and processes text information in insurance documents

Document Digitization

Document OCR

Converts text in scanned documents or images into editable text

base_model: erax-ai/EraX-VL-2B-V1.5 language:

vi
en
zh library_name: transformers license: apache-2.0 quantized_by: mradermacher tags:
erax
multimodal
erax-vl-2B
insurance
ocr
vietnamese
bcg
image-to-text
image-text-to-text

About

weighted/imatrix quants of https://huggingface.co/erax-ai/EraX-VL-2B-V1.5

static quants are available at https://huggingface.co/mradermacher/EraX-VL-2B-V1.5-GGUF

Usage

If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.

Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Link	Type	Size/GB	Notes
GGUF	i1-IQ1_S	0.5	for the desperate
GGUF	i1-IQ1_M	0.6	mostly desperate
GGUF	i1-IQ2_XXS	0.6
GGUF	i1-IQ2_XS	0.7
GGUF	i1-IQ2_S	0.7
GGUF	i1-IQ2_M	0.7
GGUF	i1-Q2_K_S	0.7	very low quality
GGUF	i1-IQ3_XXS	0.8	lower quality
GGUF	i1-Q2_K	0.8	IQ3_XXS probably better
GGUF	i1-IQ3_XS	0.8
GGUF	i1-Q3_K_S	0.9	IQ3_XS probably better
GGUF	i1-IQ3_S	0.9	beats Q3_K*
GGUF	i1-IQ3_M	0.9
GGUF	i1-Q3_K_M	0.9	IQ3_S probably better
GGUF	i1-Q3_K_L	1.0	IQ3_M probably better
GGUF	i1-IQ4_XS	1.0
GGUF	i1-IQ4_NL	1.0	prefer IQ4_XS
GGUF	i1-Q4_0	1.0	fast, low quality
GGUF	i1-Q4_K_S	1.0	optimal size/speed/quality
GGUF	i1-Q4_K_M	1.1	fast, recommended
GGUF	i1-Q4_1	1.1
GGUF	i1-Q5_K_S	1.2
GGUF	i1-Q5_K_M	1.2
GGUF	i1-Q6_K	1.4	practically like static Q6_K

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

FAQ / Model Request

See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.

Thanks

I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご