Views
No views yet
This is a LoRA adapter (~1.1 GB) — it must be loaded on top of the base modelQwen/Qwen3-VL-8B-Thinking.✅ Advantage: Lightweight — download only ~1.1 GB instead of ~17 GB.⚠️ Requirement: You must separately loadQwen/Qwen3-VL-8B-Thinking(base model, ~17 GB) first.💾 If you prefer a standalone ready-to-use model, see the merged version: E27085921/HIKARI-Deneb-8B-SkinDx-Cascade (~17 GB)
1from peft import PeftModel
2from transformers import Qwen3VLForConditionalGeneration, AutoProcessor
3import torch
4from PIL import Image
5
6# Step 1: Load base model (Qwen3-VL-8B-Thinking, ~17 GB)
7base = Qwen3VLForConditionalGeneration.from_pretrained(
8 "Qwen/Qwen3-VL-8B-Thinking",
9 torch_dtype=torch.bfloat16,
10 device_map="auto",
11 trust_remote_code=True,
12)
13
14# Step 2: Apply LoRA adapter (~1.1 GB)
15model = PeftModel.from_pretrained(base, "E27085921/HIKARI-Deneb-8B-SkinDx-Cascade-LoRA")
16processor = AutoProcessor.from_pretrained("E27085921/HIKARI-Deneb-8B-SkinDx-Cascade-LoRA", trust_remote_code=True)
17
18# Step 3: Inference — see full examples at E27085921/HIKARI-Deneb-8B-SkinDx-Cascade
19image = Image.open("skin_lesion.jpg").convert("RGB")1@misc{hikari2026,
2 title = {HIKARI: RAG-in-Training for Skin Disease Diagnosis
3 with Cascaded Vision-Language Models},
4 author = {Watin Promfiy and Pawitra Boonprasart},
5 year = {2026},
6 institution = {King Mongkut's Institute of Technology Ladkrabang,
7 Department of Information Technology, Bangkok, Thailand}
8}