Downloads · 30 days
73
10% of all-time downloads
Vintern-1B-v3_5-GGUF
ReminiScenceAI/Vintern-1B-v3_5-GGUF
Vintern-1B-v3_5-GGUF is a image-text-to-text model from ReminiScenceAI. 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 gguf. The card lists the license as apache-2.0.
This repository contains GGUF quantized weights for Vintern-1B-v3.5, developed by 5CD-AI.
All-time downloads
744
Public
Repo size
4.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf4.2 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- gguf
- License
- apache-2.0
- Access
- Public
- Created
- Apr 14, 2026
- Updated
- Apr 15, 2026
- SHA
- e3252284
Model card
From the Hugging Face model README
Vintern-1B-v3.5 GGUF
Description
This repository contains GGUF quantized weights for Vintern-1B-v3.5, developed by 5CD-AI.
Vintern-1B-v3.5 is a state-of-the-art Multimodal Large Language Model (MLLM) optimized for the Vietnamese language. Despite its compact size of 1 billion parameters, it demonstrates exceptional performance in document understanding, OCR, and detailed image description, making it ideal for edge computing and local deployment.
Key Improvements in v3.5
- Superior Vietnamese Support: Fine-tuned to understand Vietnamese cultural nuances and complex linguistic structures.
- Efficient Architecture: Based on a lightweight backbone, offering high-speed inference without compromising visual reasoning.
- GGUF Compatibility: Optimized for local execution via
llama.cpp, LM Studio, and other GGUF-supported ecosystems.
Available Quantization Methods
| File | Quantization | Description |
|---|---|---|
vintern-1b-v3.5-q4_k_m.gguf | Q4_K_M | Recommended. Balanced performance and accuracy. |
vintern-1b-v3.5-q8_0.gguf | Q8_0 | High precision, near-original quality but larger file size. |
vintern-1b-v3.5-q5_k_m.gguf | Q5_K_M | Better accuracy than Q4 with a slight increase in RAM usage. |
How to Use
1. Requirements
To run this Vision-Language Model, you need both the Main GGUF file and the Multi-Modal Projector (mmproj) file.
2. Using llama.cpp
Download the latest version of llama.cpp, following instruction from Llama.cpp's GitHub and run:
./llama-cli -m vintern-1b-v3.5-q4_k_m.gguf \
--mmproj vintern-1b-v3.5-mmproj-f16.gguf \
--image path/to/your/image.jpg \
-p "Describe this image in detail in Vietnamese."
- Task
- Image-Text-to-Text
- Library
- gguf
- License
- apache-2.0
- Languages
- vi, en
- Created
- Apr 14, 2026
- Updated
- Apr 15, 2026