PrunaAI/p-image-pixel-art-lora

Pixel Art Lora

<Gallery />

Model description

This is a text-to-image LoRA. Trigger word: use pixel art style in your prompts for best results. The gallery above shows sample generations.

Trigger word

Prepend `pixel art style` to your prompt, for example:

pixel art style <your prompt here>

Gallery

Example 1

[image]

Prompt: pixel art style A character with spiky hair and a red jacket standing in a neon-lit city street at night.

Example 2

[image]

Prompt: pixel art style A fantasy castle on a hill with a rainbow in the sky and birds flying.

Example 3

[image]

Prompt: pixel art style A cat sitting on a windowsill with plants and a cup of coffee.

Example 4

[image]

Prompt: pixel art style A robot and a human shaking hands in a futuristic laboratory.

Example 5

[image]

Prompt: pixel art style A cozy cabin in a snowy forest with smoke from the chimney.

Example prompts

Prompts used for the gallery above:

• —pixel art style A character with spiky hair and a red jacket standing in a neon-lit city street at night.
• —pixel art style A fantasy castle on a hill with a rainbow in the sky and birds flying.
• —pixel art style A cat sitting on a windowsill with plants and a cup of coffee.
• —pixel art style A robot and a human shaking hands in a futuristic laboratory.
• —pixel art style A cozy cabin in a snowy forest with smoke from the chimney.

LoRA scale

The gallery was generated with `lora_scale=0.5`. You can tune this when running inference (e.g. lower for a subtler effect, higher for a stronger style).

Use on Replicate

import replicate

output = replicate.run(
    "prunaai/p-image-lora:393190dc623ef136356d6a7221c561f48771f39c5cff90ef903ca49d1bd47574",
    input={
        "prompt": "pixel art style your prompt here",
        "lora_weights": "https://huggingface.co/davidberenstein1957/p-image-pixel-art-lora/resolve/main/weights.safetensors",
        "lora_scale": 0.5,
        "hf_api_token": "your-hf-token",
    }
)