Downloads · 30 days
7
13% of all-time downloads
RvLionXz/mufa-ai-model
mufa-ai-model is a text-to-image model from RvLionXz. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as mit.
mufa.ai adalah model Generative AI untuk menghasilkan gambar digital (text-to-image) yang dikembangkan dengan teknik model merging menggunakan metode Weighted Sum.
Downloads · 30 days
7
13% of all-time downloads
All-time downloads
53
Public
Parameters
860M
2.7 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors2.7 GB · 100%
From the Hugging Face model README
mufa.ai adalah model Generative AI untuk menghasilkan gambar digital (text-to-image) yang dikembangkan dengan teknik model merging menggunakan metode Weighted Sum.
Model ini dikembangkan sebagai objek studi kasus dalam penelitian:
"Studi Kasus Penggunaan Kecerdasan Buatan dalam Pembuatan Gambar Digital: Analisis Kemampuan dan Keterbatasan Model mufa.ai"
| Komponen | Keterangan |
|---|---|
| Metode | Weighted Sum (Model Merging) |
| Base Model | Stable Diffusion v1.5 (40%) |
| Style Model | Realistic Vision V5.1 (60%) |
| Format | float16 (diffusers) |
| Arsitektur | Stable Diffusion 1.5 (UNet + VAE + CLIP Text Encoder) |
Perbandingan bobot:
mufa.ai = (Stable Diffusion v1.5 × 0.4) + (Realistic Vision V5.1 × 0.6)
Model ini menghasilkan gambar dengan karakteristik:
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained(
"RvLionXz/mufa-ai-model",
torch_dtype=torch.float16
)
pipe = pipe.to("cuda")
prompt = "a realistic portrait of a woman with freckles, professional photography, soft lighting, 8k"
image = pipe(prompt, num_inference_steps=30).images[0]
image.save("output.png")
import requests
API_URL = "https://api-inference.huggingface.co/models/RvLionXz/mufa-ai-model"
headers = {"Authorization": "Bearer YOUR_TOKEN"}
response = requests.post(API_URL, headers=headers, json={
"inputs": "a realistic portrait of a woman, detailed skin texture",
})
MIT