Stable Diffusion is very dependent on computer configuration. In principle, it cannot run without a NVIDIA graphics card with less than 4-6GB of memory, and it cannot control poses. However, what I have discovered today is that it can be deployed using only a browser and freely using Google Colab resources to deploy your own Stable Diffusion without any device restrictions. You can use a PC, laptop, Mac, or even an iPhone or Android phone to use it for free in the cloud.
Using Google Colab to Play Stable Diffusion for Free
First, you can use any browser to access it. Open the browser and visit the website: "https://github.com/camenduru/stable-diffusion-webui-colab". After opening the webpage, scroll down to see many scripts ending with ipynb:
Scrolling down, there are dozens of scripts created by experts that can be executed for different models. Different models refer to different painting styles (the one I like to use is the ChilloutMix model selected this time). Just click on "Stable" and it will automatically jump to the Google Colab page. The current "Stable" already includes more advanced plugins such as ControlNet:
Then you will be taken to the Google Colab page (if you don't have a Google account, you need to register first), and then click on the "OPEN IN Colab" icon on the screen to use the resources (CPU, GPU, space) of the Google host in the cloud to set up a basic virtual Stable Diffusion host. Next, you need to prepare and execute the script. On the left side of this screen, there is a "Play" button. Clicking on it will start running the script:
Before running the script, a warning will pop up. Choose "Continue running" (if you are worried, then don't play with it):
Then the original "Play" button will start scrolling, and the script will start running below. You need to wait for about ten minutes here (because Google Colab needs to download Stable Diffusion and the model from the cloud and start running, so it cannot be accelerated). After about ten minutes, 2-3 URL links will appear below this page. You can choose any of them. As long as you have this link, you can control Stable Diffusion in the cloud on any device:
Then you will see the basic interface of Stable Diffusion, where you can enter positive prompts and negative prompts. Simply put, it is what you want the AI to draw and what you don't want to appear in the artwork. This is also commonly referred to as "incantation" on the Internet. The description below is for setting the image width, height, and the number of generated images. If the generated face is distorted, you can check "Restore Face", but I personally do not recommend checking it:
Then fill in the prompts you usually use (positive prompts):
(8k, RAW photo, best quality, masterpiece:1.2), (realistic, photo-realistic:1.4), ultra-detailed, (Kpop idol), perfect detail, looking at viewer, make up,
Negative prompts:
paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, bad anatomy, (long hair:1.4), DeepNegative, (fat:1.2), facing away, looking away, tilted head, {Multiple people}, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worstquality, low quality, normal quality, jpegartifacts, signature, watermark, username, blurry, bad feet, cropped, poorly drawn hands, poorly drawn face, mutation, deformed, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, extra fingers, fewer digits, extra limbs, extra arms, extra legs, malformed limbs, fused fingers, too many fingers, long neck, cross-eyed, mutated hands, polar lowres, bad body, bad proportions, gross proportions, text, error, missing fingers, missing arms, missing legs, extra digit, extra arms, extra leg, extra foot,
Then click "Generate" to generate the image:
Click on any image to display the full-size image. You can right-click and save it to your computer, or download it to your computer through the "Download" option below. Because our Google Colab is free, the images will only be saved for a few hours, and even the URL of this host will become invalid after a period of inactivity, so if you want to save it, you need to act quickly.
Below are some AI-generated photos: