lilekitty / kaneko-gen

Generates unrestricted images from text prompts using a fine-tuned Stable Diffusion model

  • Public
  • 1.1K runs
  • License
Iterate in playground

Run time and cost

This model costs approximately $0.0029 to run on Replicate, or 344 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia H100 GPU hardware. Predictions typically complete within 2 seconds.

Readme

Kaneko Gen

Kaneko Gen is a fine-tuned version of the Stable Diffusion model, modified to reduce censorship and safety filters. It enables unrestricted text-to-image generation and is intended for research, prompt testing, and creative experimentation.

Warning: This model may produce NSFW or sensitive content. Users must be 18 years or older.

Features

  • Based on Stable Diffusion (v1.4 or compatible)
  • Reduced or removed content filters
  • Supports custom prompts for open-ended image generation
  • Intended for research and experimentation only

How to Use

You can run this model on Replicate via the web interface or using Python code.

Python Example

import replicate

output = replicate.run(
    "lilekitty/kaneko-gen",
    input={
        "prompt": "a mysterious catgirl in a cyberpunk city, neon lights, ultra-detailed",
        "num_inference_steps": 30,
        "guidance_scale": 7.5
    }
)

print(output)  # This will be a URL to the generated image