---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- output:
    url: images/file_00000000c6f471f49f6d47ea454afa50.png
  text: >-
    ultra-realistic cinematic portrait of a powerful woman, luxury gothic
    fashion, black leather coat, dramatic lighting, shallow depth of field,
    highly detailed skin, realistic hair strands, 85mm lens, editorial
    photography, masterpiece, HDR, volumetric light, dark elegant atmosphere
  parameters:
    negative_prompt: >-
      low quality, worst quality, blurry, jpeg artifacts, deformed anatomy, bad
      hands, extra fingers, missing fingers, mutated hands, duplicate limbs,
      extra limbs, crossed eyes, poorly drawn face, text, watermark, logo,
      signature, oversaturated, cartoon, anime, CGI, plastic skin
- output:
    url: images/lola.png
  text: '-'
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: myanime, mystyle_v2, kawastyle
license: apache-2.0
---
# lilo

<Gallery />

## Model description 

📦 ModelHub Uploader — Universal AI Model Upload &amp; Management System

A lightweight upload engine for LoRA weights, checkpoints, embeddings, 
and quantized models. Auto-detects model type from .safetensors headers, 
generates promtchen prompt templates, and syncs to HF Hub.

Includes:
• Safetensors parser (auto-detect LoRA&#x2F;checkpoint&#x2F;embedding&#x2F;VAE)
• Promtchen JSON generator (prompt templates with placeholders)
• Gallery system with generation metadata
• HF CLI sync script
• React upload component

Usage: unzip → npm install → npm run dev → upload your weights

Supports base models: FLUX.1, SD 1.5, SDXL, Ideogram, Krea, Tongyi


## Trigger words

You should use `myanime` to trigger the image generation.

You should use `mystyle_v2` to trigger the image generation.

You should use `kawastyle` to trigger the image generation.


## Download model


[Download](/S-Jade/lilo/tree/main) them in the Files & versions tab.