imactool/hyperf-stable-diffusion

基于 rulilg/laravel-stable-diffusion 包直接平移并适配了 hyperf 框架。

我对它进行了一些改造，大部分功能保持了相同。在这里感谢一下 RuliLG ，实现了如此强大好用的 stable-diffusion 组件。

基于 Replicate API 和 stablediffusionapi 的 Stable Diffusion 实现。

• 🎨 Built-in prompt helper to create better images
• 🚀 Store the results in your database
• 🎇 Generate multiple images in the same API call
• 💯 Supports both (text to image) and (image to image)

鸣谢：原作：RuliLG，特此鸣谢!

安装

composer require imactool/hyperf-stable-diffusion

注意 表新增了platform平台字段。-- 后续会使用迁移增加，目前不考虑 :)

发布配置（包含配置文件和迁移文件)

php bin/hyperf.php vendor:publish imactool/hyperf-stable-diffusion
php bin/hyperf.php migrate

至此，配置完成。

return [
    'url' => env('REPLICATE_URL', 'https://api.replicate.com/v1/predictions'),
    'token' => env('REPLICATE_TOKEN'),
    'version' => env('REPLICATE_STABLEDIFFUSION_VERSION', 'db21e45d3f7023abc2a46ee38a23973f6dce16bb082a930b0c49861f96d1e5bf'),
];

然后在 .env 里增加 Replicate 的 token 到 REPLICATE_TOKEN即可。

使用

文字生成图片（Text to Image)

use Imactool\HyperfStableDiffusion\Prompt;
use Imactool\HyperfStableDiffusion\Replicate;

   $result = Replicate::make()->withPrompt(
            Prompt::make()
                ->with('a panda sitting on the streets of New York after a long day of walking')
                ->photograph()
                ->resolution4k()
                ->trendingOnArtStation()
                ->highlyDetailed()
                ->dramaticLighting()
                ->octaneRender()
        )->inputParams('num_outputs', 1)
        ->generate(); //根据不同的模型来传递生成的张数

图片生成图片(Image to Image)

use Imactool\HyperfStableDiffusion\Prompt;
use Imactool\HyperfStableDiffusion\Replicate;
use Intervention\Image\ImageManager;

//这里使用了 intervention/image 扩展来处理图片文件，你也可以更换为其他的
 $sourceImg =  (string) (new ImageManager(['driver' => 'imagick']))->make('path/image/source.png')->encode('data-url');

$prompt = 'Petite 21-year-old Caucasian female gamer streaming from her bedroom with pastel pink pigtails and gaming gear. Dynamic and engaging image inspired by colorful LED lights and the energy of Twitch culture, in 1920x1080 resolution.';
$result = Replicate::make()
    ->converVersion('a991dcab77024471af6a89ef758d98d1a54c5a25fc52a06ccfd7754b7ad04b35')
    ->withPrompt(
        Prompt::make()
            ->with($prompt)
    )
    ->inputParams('image',$sourceImg)
    ->inputParams('negative_prompt', 'disfigured, kitsch, ugly, oversaturated, greain, low-res, Deformed, blurry, bad anatomy, disfigured, poorly drawn face, mutation, mutated, extra limb, ugly, poorly drawn hands, missing limb, blurry, floating limbs, disconnected limbs, malformed hands, blur, out of focus, long neck, long body, ugly, disgusting, poorly drawn, childish, mutilated, mangled, old, surreal, calligraphy, sign, writing, watermark, text, body out of frame, extra legs, extra arms, extra feet, out of frame, poorly drawn feet, cross-eye, blurry, bad anatomy')
    ->inputParams('strength', 0.5)
    ->inputParams('upscale', 2)
    ->inputParams('num_inference_steps', 25)
    ->inputParams('guidance_scale', 7.5)
    ->inputParams('scheduler', 'EulerAncestralDiscrete')
    ->inputParams('num_outputs', 1)
    ->generate();

查询结果

use Imactool\HyperfStableDiffusion\Replicate;
 $freshResults = Replicate::get($replicate_id);

Generating prompts

There are several styles already built-in:

Method	Prompt modification
realistic()	{prompt}, realistic
hyperrealistic()	{prompt}, hyperrealistic
conceptArt()	{prompt}, concept art
abstractArt()	{prompt}, abstract art
oilPainting()	{prompt}, oil painting
watercolor()	{prompt}, watercolor
acrylic()	{prompt}, acrylic
pencilDrawing()	{prompt}, pencil drawing
digitalPainting()	{prompt}, digital painting
penDrawing()	{prompt}, pen drawing
charcoalDrawing()	{prompt}, charcoal drawing
byPicasso()	{prompt}, by Pablo Picasso
byVanGogh()	{prompt}, by Vincent Van Gogh
byRembrandt()	{prompt}, by Rembrandt
byMunch()	{prompt}, by Edvard Munch
byKlimt()	{prompt}, by Paul Klimt
byKandinsky()	{prompt}, by Jackson Pollock
byMonet()	{prompt}, by Claude Monet
byDali()	{prompt}, by Salvador Dali
byDegas()	{prompt}, by Edgar Degas
byKahlo()	{prompt}, by Frida Kahlo
byCezanne()	{prompt}, by Pablo Cezanne
photograph()	a photo of {prompt}
highlyDetailed()	{prompt}, highly detailed
surrealism()	{prompt}, surrealism
trendingOnArtStation()	{prompt}, trending on art station
triadicColorScheme()	{prompt}, triadic color scheme
smooth()	{prompt}, smooth
sharpFocus()	{prompt}, sharp focus
matte()	{prompt}, matte
elegant()	{prompt}, elegant
theMostBeautifulImageEverSeen()	{prompt}, the most beautiful image ever seen
illustration()	{prompt}, illustration
digitalPaint()	{prompt}, digital paint
dark()	{prompt}, dark
gloomy()	{prompt}, gloomy
octaneRender()	{prompt}, octane render
resolution8k()	{prompt}, 8k
resolution4k()	{prompt}, 4k
washedColors()	{prompt}, washed colors
sharp()	{prompt}, sharp
dramaticLighting()	{prompt}, dramatic lighting
beautiful()	{prompt}, beautiful
postProcessing()	{prompt}, post processing
pictureOfTheDay()	{prompt}, picture of the day
ambientLighting()	{prompt}, ambient lighting
epicComposition()	{prompt}, epic composition

Additionally, you can add custom styles with the following methods:

• as(string $canvas): to add a string at the beginning (i.e. "a photograph of")
• paintingStyle(string $style): to add a painting style (i.e. realistic, hiperrealistic, etc.)
• by(string $author): to instruct the system to paint it with the style of a certain author
• effect(string $effect): to add a finishing touch to the prompt. You can add as many as you want.

To learn more on how to build prompts for Stable Diffusion, please enter this link.

基于 stablediffusionapi 平台 https://stablediffusionapi.com/docs/

或者 Postman Collection

文字生成图片（Text to Image)

use Imactool\HyperfStableDiffusion\StableDiffusion;

 $res = StableDiffusion::make()
                    ->useDreamboothApiV4()
                    ->withPayload('key', '')
                    ->withPayload('model_id', 'anything-v4')
                    ->withPayload('prompt', 'ultra realistic close up portrait ((beautiful pale cyberpunk female with heavy black eyeliner)), blue eyes, shaved side haircut, hyper detail, cinematic lighting, magic neon, dark red city, Canon EOS R3, nikon, f/1.4, ISO 200, 1/160s, 8K, RAW, unedited, symmetrical balance, in-frame, 8K')
                    ->withPayload('negative_prompt', 'painting, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, deformed, ugly, blurry, bad anatomy, bad proportions, extra limbs, cloned face, skinny, glitchy, double torso, extra arms, extra hands, mangled fingers, missing lips, ugly face, distorted face, extra legs, anime')
                    ->withPayload('width', '512')
                    ->withPayload('height', '512')
                    ->withPayload('samples', '1')
                    ->withPayload('num_inference_steps', '30')
                    ->withPayload('seed', null)
                    ->withPayload('guidance_scale', '7.5')
                    ->withPayload('webhook', null)
                    ->withPayload('track_id', null)
                    ->text2img();

                var_dump( $res);

License

The MIT License (MIT). Please see License File for more information.