Downloads · 30 days
6
3% of all-time downloads
tuanio/ft-moellava-qwen1.5-1.8b-vista-lora-2ep
ft-moellava-qwen1.5-1.8b-vista-lora-2ep is a text classification model from tuanio. Use it when you need a label for a piece of text. It is set up for adapter-transformers. The card lists the license as mit.
<p align="center" <div style="display: flex;text-align: center;" <div <img src="https://firebasestorage.googleapis.com/v0/b/database-7ca5c.appspot.com/o/llm%2F68747470733a2f2f7331312e617831782e636f6d2f323032332f31322f…
Downloads · 30 days
6
3% of all-time downloads
All-time downloads
173
Public
Repo size
480 MB
Likes
2
Public
Click a slice to open those files.
.safetensors240 MB · 98%
From the Hugging Face model README
We are excited to present MoE-LLaVA-Qwen1.5-1.8B×4-Top2, tailored for the Vietnamese language. This model is part of our ongoing efforts to develop Vision Language Models (VLM) for Vietnamese, a domain that is currently limited and predominantly features larger models (~7B parameters). Our model activates approximately 2.2B 🤗😎 parameters per call, significantly reducing the memory footprint, and it can be quantized for local execution.
Our model is trained on the comprehensive Vi-VLM/Vista dataset, which includes around 700,000 Vietnamese vision-language samples curated by Gemini Pro. We employed various prompt engineering techniques, including:
For the COCO dataset, we utilized Llava-style prompts to generate data. For the ShareGPT4V dataset, translation prompts were applied.
The dataset may contain biases originating from its sources. Users should remain aware of these potential biases when utilizing the dataset.
This dataset represents the first stage of a two-stage development process for a larger model. Stay tuned for future developments by subscribing to our updates.