Downloads · 30 days
7
9% of all-time downloads
gemma3-4b-distil-ft
AhmetSemih/gemma3-4b-distil-ft
gemma3-4b-distil-ft is a image-text-to-text model from AhmetSemih. 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. The card lists the license as apache-2.0.
This model is a fine-tuned version of AhmetSemih/gemma3-4b-freezed-pretrainfinal, specifically optimized for Turkish logic and reasoning tasks.
All-time downloads
82
Public
Parameters
4B
7.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors7.9 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- License
- apache-2.0
- Model type
- gemma3
- Access
- Public
- Created
- Mar 3, 2026
- Updated
- Mar 13, 2026
- SHA
- 18b3757e
Base models
Model card
From the Hugging Face model README
Uploaded model
Gemma 3 4B Turkish Reasoning (Distilled from Pre-trained)
This model is a fine-tuned version of AhmetSemih/gemma3-4b-freezed-pretrain_final, specifically optimized for Turkish logic and reasoning tasks.
🧩 Key Features
- Turkish Linguistic Foundation: Leverages the custom-pretrained embeddings from the parent model for superior natural language processing in Turkish.
- Logical Reasoning: Automatically breaks down complex questions into a "Thinking Process" before arriving at an answer.
- Instruction Following: Trained to adhere to a structured prompt format for consistent multi-choice question (MCQ) performance.
📝 Prompt Template
To trigger the model's reasoning mode, use the following structure:
### Soru:
### Seçenekler:
[A) ... B) ... C) ...]
### Düşünce Süreci:
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- gemma3
- License
- apache-2.0
- Languages
- tr
- Created
- Mar 3, 2026
- Updated
- Mar 13, 2026