TinyLLaVA-SmolLM2-135M-Instruct-dinov2-small-finetuned

sbrzz/TinyLLaVA-SmolLM2-135M-Instruct-dinov2-small-finetuned

TinyLLaVA-SmolLM2-135M-Instruct-dinov2-small-finetuned is a visual question answering model from sbrzz. Use it for the visual question answering task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

Downloads · 30 days

7% of all-time downloads

All-time downloads

226

Public

Parameters

157M

314 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors314 MB · 100%

At a glance

Task: Visual Question Answering
License: apache-2.0
Model type: tinyllava
Access: Public
Created: Nov 14, 2024
Updated: Nov 14, 2024
SHA: bf815de8

Base models

Model card

From the Hugging Face model README

Finetuned, 500 epochs

MMMU

Category	# Samples	Accuracy
Overall	900	0.250
Overall-Art and Design	120	0.217
Art	30	0.167
Art Theory	30	0.267
Design	30	0.300
Music	30	0.133
Overall-Business	150	0.300
Accounting	30	0.367
Economics	30	0.300
Finance	30	0.300
Management	30	0.267
Marketing	30	0.267
Overall-Science	150	0.227
Biology	30	0.200
Chemistry	30	0.167
Geography	30	0.267
Math	30	0.267
Physics	30	0.233
Overall-Health and Medicine	150	0.293
Basic Medical Science	30	0.400
Clinical Medicine	30	0.167
Diagnostics and Laboratory Med.	30	0.267
Pharmacy	30	0.300
Public Health	30	0.333
Overall-Humanities and Soc. Sci.	120	0.267
History	30	0.167
Literature	30	0.233
Sociology	30	0.367
Psychology	30	0.300
Overall-Tech and Engineering	210	0.210
Agriculture	30	0.200
Architecture and Engineering	30	0.200
Computer Science	30	0.200
Electronics	30	0.167
Energy and Power	30	0.167
Materials	30	0.267
Mechanical Engineering	30	0.267

Task: Visual Question Answering
Type: tinyllava
License: apache-2.0
Languages: en
Created: Nov 14, 2024
Updated: Nov 14, 2024