Mini Journey

Introduction

I created Mini Journey as an attempt to replicate midjourney style, this model combines various AI models, both SFW and NSFW, with carefully selected weights at each stage of the mixing process.

To be honest, the model itself doesn't really have perfect prompt adherence, you can check it on the sample section. But boy, the results created by this model is amazing!

Best For: Characters, Digital Illustrations

Bad For: Realistic, Photography, Scenery (will often include a character)

Step by Step - How to use Mini Journey via AiSudo

1. Open the AISudo app on your device and go to the main page.
2. Click the "+ New Session" button.
3. On the next page, click "Generate AI Art" to start a new art generation session.
4. Now, choose "Mini Journey" from the list of available AI models. This is the model that will turn your text into an image.
5. Click "Create Session" to start your new session. This will take you to the main creation page where you can enter your text and generate an image.

Cost: 2-5 credits per generation, depends on number of steps and image size.

Sample Images

These are some examples of images generated using Mini Journey. You can recreate these images by copying the prompts below and pasting them into the prompt field in the AISudo app. The sample generated with size 768x768 and 20 inference steps.

1girl, solo, looking at viewer, blue eyes, black hair, hood, armor, hood up

1boy, solo, smile, male_focus, grin, bandages

stars, nebula, iridescent clouds, reflective water, soft starlight, ethereal glow, detailed woman, flowing white dress, serene expression

floral shirt, white dress, sunset lighting, bokeh, shallow depth of field, skin texture, warm tones, gold jewelry, ferris wheel, carnival lights

sunflowers, vibrant yellow, textured straw hats, colorful dresses, detailed faces, golden light, oil painting, impasto technique, brushstrokes, pastoral scene, serene

outdoors, sky, cloud, tree, no humans, grass, cloudy sky, storm, single tree, haystack, tornadoes

flames, intense fire, burning sign, red brick storefront, shattered windows, rain-slicked pavement, reflective pools, smoke, black smoke plume, emergency lights, chaotic scene

shadowy figure, feathered cloak, glowing eyes, dramatic lighting, dark bedroom, textured fabric, ominous presence, candle light, low-angle perspective

female knight, ornate gold armor, intricate embroidery, flowing white hair, moonlit forest, glowing embers, ethereal light, sharp details, polished metal, dynamic pose, cinematic lighting

1boy, gloves, standing, weapon, holding weapon, boots, weapon belt, tree, holding gun, tree stump, brown eyes

Parameters

You can customize the output by tweaking various parameters, giving you a high level of flexibility to fit your image generation needs.

Basic Settings

• Inference Steps : This setting controls how many steps the AI model takes to generate the image. A higher number of steps can result in more detailed images, but it also takes longer to generate (and more expensive).
• Image Width: This setting lets you choose how wide you want the generated image to be.
• Image Height: This setting lets you choose how tall you want the generated image to be.

Advanced Settings

• Scheduler : The scheduler dictates how the noise reduction happens over each iteration, impacting the quality, speed, and style of the generated image.
• Seed : This is a random number that the AI model uses to start generating the image. If you use the same seed with the same text, you'll get the same image. If you want to generate a different image with the same text, try changing the seed.
• Guidance Scale : The guidance scale controls how closely the generated images match the prompt. Higher values result in images more similar to the prompt, while lower values give the AI model more freedom.
• Clip Skip : Clip Skip is a feature that skips part of the image generation process, leading to slightly different results. This means the image renders faster, too.
• Censor NSFW : This parameter is used to enable or disable the NSFW censor.

Best setup for Mini Journey:

• Steps: 20-40, you can go 50 if you need more detail.
• Guidance/CFG scale: 5-10
• Scheduler: Euler Ancestral or DPM++ 2M
• Clip Skip: 2

Try AiSudo now, and get free 15 daily credits!