Downloads · 30 days
0
AD-Styles/mini-llava-v4
mini-llava-v4 is a machine learning model from AD-Styles. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.
처음부터 조립한 멀티모달 LLM (vlm-from-scratch-v4) 의 학습된 가중치.
Downloads · 30 days
0
Access
Public
Updated May 16, 2026
Repo size
176 MB
Likes
0
Public
Click a slice to open those files.
.safetensors73.9 MB · 84%
From the Hugging Face model README
처음부터 조립한 멀티모달 LLM (vlm-from-scratch-v4) 의 학습된 가중치.
| 파일 | 설명 |
|---|---|
projector.pt | MultiModalProjector (CLIP 768 → LLM 1536) state_dict |
lora_adapter/ | Qwen2.5-1.5B 전 linear layer LoRA 어댑터 (r=16) |
<image> 토큰으로 Qwen2.5 내장 <|image_pad|> 를 재사용하므로 adapter 에
embedding 군더더기가 없다 (70 MB 전부 LoRA).
추론 코드는 github.com/AD-Styles/vlm-from-scratch-v4
의 src/ 참고. 데모: HF Space AD-Styles/mini-llava-v4-demo.