Dreamshaper v7 Guide

Dreamshaper v7 model is fine-tuned for a portrait illustration style that lies between photorealistic and computer graphics. It is easy to use. It can also handle many other styles. You can think of it as a Midjourney in Stable Diffusion.

What This Model Excels At

  • Crafting photography/3D render/Octane style images.
  • Generating SFW/NSFW images.
  • Crafting portrait/illustration/drawing/sketching style arts.
  • Generating woman/man human(humanoid) images.
  • Generating non-human object/landscape images.
  • Generating fantasy/sci-fi themed images.
  • Great lighting/shadow usage.
  • Great depth/blur usage.
  • Recognizes most art/artist styles.

What To Be Aware of About This Model

  • Crafting anime style with "(anime coloring, anime screencap)" prompt.

  • Model may struggle to maintain anime style. Often switches to 3D images.

  • The greater the distance, the lower the quality. Prefers closeup and medium shots.

  • Increases quality with " trending on .... , award winning art, uhd, 8k" like prompts.

  • Using 8k with photo/portrait is better than single usage.

  • Do not use FaceFix if you will not make a realistic image. It ruins image.

  • Use sharp/soft focus and depth of field to make more realistic images.

  • Using "blurry background/foreground, bokeh" like terms can increase quality of main subject.

Prompting Style

Dreamshaper prompts use a many unique terms varying from photography to sketching and beyond.

  • Steps: 20-30
  • CFG Scale: 7-12 (better at 7-9)
  • Sampler: DPM++ 2M Karras

Prompt Component Order

  1. Art/Artist names
  2. Angle
  3. Style
  4. Subject statement
  5. Quality/detail terms
  6. Major details
  7. Minor details
  8. Lighting/Shadows
  9. Additional modifiers
  10. Additional quality modifiers like "trending on deviantart"

closeup, 8k portrait of a woman standing in a forest, masterpiece, extremely detailed, best quality, highres, soft light, soft backlighting, soft shadows, (best image:1.5)

Sample Prompts

Armored Liberation Warrior

armored_warrior.png

Positive Prompt : (by frank hong, by maciej kuciara), closeup upperbody, illustration of a man in sci fi armored suit, (masterpiece, award winning art), concept art, game art, (8k, hd, best quality, highres:1.4), (extremely detailed:1.5), (octane render unreal engine), (cgi:1.2), (sci fi indoors), metallic colors, (helmet on), (steel metal, worn), sharp focus, volumetrics dtx, cinematic spotlighting, sharp shadows, (reflection of light), metal reflections, (depth of field, blurry background, blurry foreground), trending on artstation, trending on deviantart, (best image:1.5)

Negative Prompt : (worst quality:1.5), (worst image:1.5), (low detail:1.5), (low quality, bad quality), leather, deformed, (lowres), monochrome, grayscale, watermark, text, red eyes, (compression artifacts)

Right-Hand of Shogun

right_hand_of_shogun.png

Positive Prompt: : 8k portrait of mysterious samurai woman, (masterpiece, intricately detailed), (photorealistic:1.4), extreme quality, highres, digital photography, upper body, the most beautiful woman ever, outdoors, (edo period, japan), ponytail, kimono, cleavage, professional photograph of alluring woman detailed, sharp focus, ([dramatic:cinematic:0.2] lighting), octane render, volumetric dtx, uhd, trending on artstation, (film grain, blurry background, blurry foreground, bokeh, depth of field, motion blur), (best image, best quality:1.5)

Negative Prompt : (worst quality:1.5), (worst image:1.5), (low detail:1.5), (low quality, bad quality), (ugly), (deformed eyes:1.3), deformed, lowres, monochrome, grayscale, watermark, text

Sports Car on Highway

sports_car.png

Positive Prompt : 8k photo of a super car, (masterpiece, award winning art), (hd, best quality, highres:1.4), (extremely detailed:1.5), (octane render unreal engine), (cgi:1.2), (on highway, mountains), asphalt ground, metallic colors, volumetrics dtx, ambient lighting, sharp shadows, (reflection of light), metal reflections, (depth of field, blurry background, blurry foreground), trending on artstation, trending on deviantart, (best image:1.5)

Negative Prompt :(worst quality:1.5), (worst image:1.5), (low detail:1.5), (low quality, bad quality), deformed, (lowres), monochrome, grayscale, watermark, text, (compression artifacts)

Anime Style

anime_style.png

Positive Prompt : (anime coloring, anime screencap), 1woman, white gown, upper body

Negative Prompt :(worst quality:1.5), (worst image:1.5), (low detail:1.5), (low quality, bad quality), 3d, render, deformed, (lowres), monochrome, grayscale, watermark, text, (compression artifacts), red eyes

Prompt Style Comparison

Below you can see the difference in generation quality between applying only the general prompt style in Prompting 101 and applying the recommended style specific to this model.

ResultPrompt
dreamshaperv7_standard.pngcloseup portrait shot, a woman standing in a forest, masterpiece, extremely detailed, soft light, soft shadows, soft backlighting, (best image,best quality:1.5)

Final Thoughts

DreamShaper v7 is an excellent model, especially for beginner users. It is easy to use and can create many different styles. It also works well with both short and long prompts. It has a generation capacity from the most realistic art to the most basic drawing with the right prompts. If you don't know where to start AI art, DreamShaper will be a perfect choice for you. If you are already skilled, it can make all the dreams you have ever imagined come true.