Models

SD 1.4 (Stable Diffusion)

Stable Diffusion is a latent text-to-image diffusion model capable of generating photo-realistic images given any text input. Prodia's main model is the model version 1.4.

It is the most general model on the Prodia platform however it requires prompt engineering for great outputs.

Here are a few examples of the prompt close-up of woman indoors, wide-angle, soft lighting, style of polaroid

The reason for the distorted faces is because the model requires more input data to generate better results. By altering the prompt to be more specific and adding words such as realistic, highly detailed, art-station, trending, masterpiece, realistic face, realistic skin, detailed eyes, great artwork to the end of the prompt you can improve it to results below:

Anything V3 and 4.5

These are very strong Anime models.

Anything V3 is more versatile, but some users on the Prodia Discord reports that generations are generally more gray in color.

Anything V4.5 is very good for high-quality anime, but follows one style more strictly. You can generate great quality images with simpler prompts.

Prompt

Dorohedoro style Wired Wonder - Enter the Fascinating World of a Male Made Entirely of Wires, Cables, and Threads in this Striking, Adorned with wires, cable-suit, (Sitting down rooftop close up), Panoramic, Psychedelic, With Mind-Blowing Animation Techniques, including Intricate Character Designs and Hypnotic Backgrounds of Old Metropole, the Creative Vision of Acclaimed Artists Takehiko Inoue, Tom bagshaw, James Jean, Kengo Hanazawa, Haro Aso, and yuko shimizu Unfolds in a Stunning Display of Cybernetic Artistry

Negative Prompt

canvas frame, cartoon, 3d, ((disfigured)), ((bad art)), ((deformed)),((extra limbs)),((close up)),((b&w)), weird colors, blurry, (((duplicate))), ((morbid)), ((mutilated)), [out of frame], extra fingers, mutated hands, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers), (((long neck))), Photoshop, video game, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, mutation, mutated, extra limbs, extra legs, extra arms, disfigured, deformed, cross-eye, body out of frame, blurry, bad art, bad anatomy, 3d render

Model: Anything V4.5
Seed: -1
CFG Scale: 7
Steps: 30

Analog V1

Trained on analog photographs. Good for photography and realistic humans. Very good at celebs.

This model also requires many inputs and a fair amount of negative prompts.

Prompt

(masterpiece:1.3), (best quality), (highres:1.1), ((absurdres)), (incredibly detailed), wide angle, raw photo of emilia clarke as wonder woman, blonde hair, ultra quality, hd, wet hair, detailed eyes, high quality lighting, detailed skin, skin texture, (high detailed skin:1.4)

Negative Prompt

canvas frame, cartoon, 3d, ((disfigured)), ((bad art)), ((deformed)),((extra limbs)),((close up)),((b&w)), weird colors, blurry, (((duplicate))), ((morbid)), ((mutilated)), [out of frame], extra fingers, mutated hands, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers), (((long neck))), Photoshop, video game, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, mutation, mutated, extra limbs, extra legs, extra arms, disfigured, deformed, cross-eye, body out of frame, blurry, bad art, bad anatomy, 3d render

Model: Analog V1
Seed: -1
CFG Scale: 4
Steps: 30

Here, the Analog V1 excels at doing celebs, but needs a lot of modifiers to keep it in line and produce something nice.

TheAlly’s Mix II

Good at painted portraits and landscapes. Does not require many prompts to generate good quality images.

Elldreth’s Vivid

Vivid, perhaps the most versatile model for easy generations, is good for realistic portraits, stylized characters, landscapes, fantasy, sci-fi, anime, and horror.

Prompt

a candid photorealistic shot of Tyrion Lannister from Game of Thrones at Winterfell, portrayed by Peter Dinklage, The Imp, mid-winter with heavy snowfall, standing in the courtyard wearing intricate Lannister attire, highly detailed, detailed face, realistic skin, detailed eyes, cinematic composition, 4K, UHD, backlit, pensive and curious expression, artstation, trending

Negative Prompt

canvas frame, cartoon, 3d, ((disfigured)), ((bad art)), ((deformed)),((extra limbs)),((close up)),((b&w)), weird colors, blurry, (((duplicate))), ((morbid)), ((mutilated)), [out of frame], extra fingers, mutated hands, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers), (((long neck))), Photoshop, video game, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, mutation, mutated, extra limbs, extra legs, extra arms, disfigured, deformed, cross-eye, body out of frame, blurry, bad art, bad anatomy, 3d render

Model: Elldreth’s Vivid
Seed: -1
CFG Scale: 6
Steps: 30