Downloads · 30 days
5
7% of all-time downloads
Gemma3-4B_WEASEL
yeonjooooni/Gemma3-4B_WEASEL
Gemma3-4B_WEASEL is a image-text-to-text model from yeonjooooni. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers.
This model is a WEASEL fine-tuned variant of google/gemma-3-4b-it for web-agent style reasoning and action generation.
All-time downloads
73
Public
Parameters
4.3B
8.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors8.6 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- Model type
- gemma3
- Access
- Public
- Created
- Jun 1, 2026
- Updated
- Jun 2, 2026
- SHA
- d4372997
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
Gemma3-4B_WEASEL
This model is a WEASEL fine-tuned variant of google/gemma-3-4b-it for web-agent style reasoning and action generation.
Model description
- Base model:
google/gemma-3-4b-it - Fine-tuning objective: web task solving with reasoning/action style supervision
- Format: Hugging Face Transformers + Safetensors
Intended use
This model is intended for research on web-agent behavior, instruction following, and reasoning traces.
Training data
Fine-tuned on an AgentTrek-style web-agent dataset.
If released separately, link it here: yeonjooooni/agenttrek-dataset
Limitations
- May generate incorrect or unsafe browser actions
- Reasoning traces do not guarantee factual correctness
- Not evaluated for safety-critical use
Loading
from transformers import AutoProcessor, AutoModelForImageTextToText
model_id = "yeonjooooni/Gemma3-4B_WEASEL"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForImageTextToText.from_pretrained(model_id)
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- gemma3
- Languages
- en
- Created
- Jun 1, 2026
- Updated
- Jun 2, 2026