How to Use Stable Diffusion Negative Prompts to Create Amazing Images

Stable diffusion negative prompt featured image

Stable diffusion is a new AI model that can generate high-quality images from text prompts. However, it can sometimes generate images that are blurry, pixelated, or otherwise low quality. 

Today I’m going to share with you a secret weapon that will take your image-creation skills to the next level: Stable Diffusion negative prompts. That’s right, negative prompts are a game-changer for image creation, and I’m going to show you why and how to use them.

What are Stable Diffusion Negative Prompts?

A negative prompt is a way to tell a text-to-image AI model what you don’t want to see in the generated image. It is a parameter that tells the model to avoid generating images that match the description of the negative prompt.

For example, you could use stable diffusion negative prompts to tell the model to avoid generating images that are blurry, pixelated, or have a certain color scheme.

In this blog article, we will discuss the basics of stable diffusion negative prompts. We will also provide a list of common negative prompts that you can use to improve the quality of your generated images.

Why Use Negative Prompts?

They can help you remove unwanted objects, styles, or abnormalities from the original generated image. For example, if you want to create an image of a woman you can use negative prompts like this: bad anatomy, disfigured features, blurry details ,cloned faces

image of woman

Negative prompts can help you explore different variations and possibilities of your image. If you want to create an image of a cyberpunk forest by Salvador Dali, but you don’t want any trees or green color in it, you can use negative prompts.

Positive Prompt : cyberpunk forest by Salvador Dali

Negative Prompt  : Green, Trees

mystical forest

If you’re trying to make a picture of a beach, but you want it to show only the shore without any people, you can achieve this by using negative prompts to take out those elements from the image. In other words, you can guide the process by telling the tool what not to include, which will help you get the picture you’re aiming for with just the beach and no human figures. This way, you can create a serene beach scene without any people in it.

Positive Prompt : Beach shore, beautiful masterpiece

Negative Prompt  : Humans, crowd, people

images of beach

Few other examples

Positive Prompt :  Cute happy Dog, best quality, depth of field, ultra high res, intricate, photorealistic, masterpiece, ultra-detailed

Negative Prompt  : semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands antd fingers, deformed, distorted, disfigured, poorly drawn, bad anatomy, wrong anatomy, mutation, mutated, ugly, disgusting

images of two dogs

Positive Prompt :  Photo of a car, uhd, dslr, soft lighting, high quality, cinematic shot, soft lighting, detailed background

Negative Prompt  : semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated ,deformed, distorted, disfigured, poorly drawn, bad anatomy, ugly, disgusting, amputation

stable diffusion negative prompts 1

Here are some tips on using negative prompts for Stable Diffusion

  • Use specific and descriptive words. Instead of using vague words like “bad” or “ugly,” use words that describe what makes something bad or ugly for you. For example, instead of saying “no bad images,” you could say “no blurry images,” “no distorted images,” or “no images with too many objects.”
  • Use multiple words. You can use multiple words in a negative prompt, separated by commas. This will help to ensure that the model does not generate any images that contain any of the words in the negative prompt. For example, you could say “no blurry, distorted, or noisy images.”
  • Use weights. You can use weights to adjust the importance of each word in a negative prompt. This can be helpful if you want to emphasize certain words or phrases. For example, you could say “no blurry images: -2.0, no distorted images: -1.0.” This would tell the model that it is more important to avoid blurry images than distorted images.
  • Experiment. The best way to learn how to use negative prompts effectively is to experiment. Try different combinations of words, weights, and phrases to see how they affect the output of the model. You may be surprised at what you can achieve!

Stable Diffusion Negative Prompts

Missing armsPoorly drawn faceTextDuplicate featuresLong neckAbsent limbs
MalformedKitschSignatureDehydratedUnnatural anglePoorly drawn face
ChildishGrainyExtra legsPoorly drawn feetUnfocusedMultiple views
Incorrect ratioAmputeeMacabreShortenedTilingGrains
Mutated handsMorbidExtra handsUncreativeUglyMistake
UnattractiveMistakeDistorted limbsBlurryIncorrect ratioBad eyes
Low qualityIncorrect physiologyDisconnected limbBad anatomyUnsightlyFlaw
Unnatural poseRevolting dimensionsBeyond the bordersExtra armsIdentifying markLowres
DuplicateOut of frameIndistinctUsernameUnnatural posePixelated
DisfiguredPoorly drawn handsWatermarkUnsightlyDouble faceBad illustration
Missing fingersOverexposedExtra fingersLow resolutionDeformedPoorly drawn hands
UnfocusedCloned faceTrimmedMissing handsExtra limbsExtra arms


Negative prompts are a powerful tool that can be used to improve the quality and consistency of the images generated by Stable Diffusion. By using negative prompts, you can tell the model what you don’t want to see in the generated images, which can help to prevent the model from generating blurry, distorted, or noisy images. You can also use negative prompts to modify the style of the generated images.

Join Our Tech Community!

Subscribe & get an instant FREE gift! + receive news, updates, and special gifts straight to your inbox.

You Might Also Like

Where Should We Send The Gift?

Provide your name and email, and we’ll send the guide directly to your inbox!

How to Create the Perfect ChatGPT Prompt for Precise Answers!

Crafting an effective prompt is a learnable skill. Your choice of words in the prompt directly influences ChatGPT’s responses. This guide will show you the key elements for getting the right response.