Quick Generate
Select a style, write what you want to see and click Generate
What you want in the image.
Clear
Random
* This model is available for use and will generate similar character to the ones seen in the images.

Model Description by Creator

RetroJoy: the joy of retro game nostalgia.

Hey, everyone, I'm back from a long hiatus with brand new LoRA styles!

This time, I've created several LoRAs based off of video game graphics from the past, and I chose beloved game consoles from the mid-late 90s: the original PlayStation (or PS1), Nintendo 64 (or N64), and even the Sega Dreamcast!

How I trained (via Kohya_ss):

PS1: Trained on 45 images and 3 epochs for a total of 13,500 steps.

N64: Trained on 43 images and 3 epochs for a total of 12,900 steps.

Dreamcast: Trained on 51 images and 2 epochs for a total of 10,200 steps.

Network Rank (dim): 128

Network Alpha: 32

Recommended settings for each LoRA:

PS1:

LoRA weight: ~0.8-1.1, but it is recommended to put weight on 1 for the full effect of the pixelated, warped 32-bit graphics of the PlayStation. Make sure you add "ps1 style" in your prompt (putting it at the end will have a more accurate effect). and adding the words "low poly" and/or "pixelated" will help out the prompt as well.

N64:

LoRA weight: ~0.8-1.3, but it is recommended to put weight on 0.8, 0.9, or 1 for the full effect of the 64-bit graphics of the Nintendo 64. Make sure you add "n64 style" in your prompt (putting it at the end will have a more accurate effect).

Dreamcast:

LoRA weight: ~0.8-1.2, but it is recommended to put weight on 0.9 or 1 for the full effect of the 128-bit graphics of the Dreamcast. Make sure you add "dreamcast style" in your prompt (putting it at the end will have a more accurate effect).

Universal settings for all LoRAs:

Recommended checkpoint: Deliberate (the latest version of it, and because it's the most accurate one that I've personally used, but you can try out others if need be)

Recommended samplers: DPM++ 2M Karras or Euler a

Adding the words "low poly" and/or "pixelated" will help out the prompts with their wobbling and warping (ESPECIALLY for the PS1).

You can use short and long prompts, but it is recommended to use shorter prompts for more accurate effects of the LoRAs.

You can use negatives as well, like this one (and EasyNegative prompts can work wonders, too): bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature,

less than five fingers, EasyNegative

Denoising (if you're planning on doing hires. fix): ~0.55-0.7

CFG Scale: 5-7

This is all for the fun and appreciation of the gaming past, so suggestions and feedback are welcome to help out with the next versions (or maybe I could add even more retro game styles)!

Enjoy!

Images Generated With This Model

Loading...

Similar Models

PixelArtRedmond - Pixel Art Loras for SD XL
V2
PixelArtRedmond - Pixel Art Loras for SD XL
(5.00)
0
8 Ratings
Makima (Chainsaw Man) Character Lora / Makima's Eyes Concept Lora
V1
Makima (Chainsaw Man) Character Lora / Makima's Eyes Concept Lora
(0.00)
0
0 Ratings
Sojourn | Overwatch
V1
Sojourn | Overwatch
(0.00)
0
0 Ratings
PixelArtRedmond 1.5V- Pixel Art Loras for Liberte Redmond SD 1.5!
V1
PixelArtRedmond 1.5V- Pixel Art Loras for Liberte Redmond SD 1.5!
(5.00)
0
3 Ratings
Pixel_Fusion_LORA
N/A
V2
Pixel_Fusion_LORA
(5.00)
0
3 Ratings
FLUX Lora XL
V2
FLUX Lora XL
(0.00)
0
0 Ratings
Project_Pinay
N/A
V1
Project_Pinay
(0.00)
0
0 Ratings
【Omega Strikers】Juliette Charecter LoRA(朱莉葉人物模組)
V1
【Omega Strikers】Juliette Charecter LoRA(朱莉葉人物模組)
(5.00)
0
4 Ratings
N64 Style
V2
N64 Style
(5.00)
0
3 Ratings
Nebula_style(星云幻想)
V1
Nebula_style(星云幻想)
(4.00)
0
1 Ratings

Train Models

Train AI Models On Your Own Pictures

You can use your own pictures to train AI models. Reimagine yourself in new ways with our AI-powered image generator.
Create pictures of yourself around the world, in your dream outfit, as a video game character, or in countless art styles. The possibilities are endless.
Train Your Own Model

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.