Better Than Midjourney: Openjourney Stable Diffusion

All Your Tech AI
6 Mar 202309:52

TLDRThe video introduces a method to run mid-journey as a local service on a PC, enabling unlimited free image generation. It guides viewers through downloading an open-source mid-journey model from Hugging Face and integrating it with a local version of stable diffusion. The host demonstrates the process using various prompts and showcases the resulting images, highlighting the potential of this approach to produce detailed, stylistic artwork without the need for a monthly subscription.

Takeaways

  • 🚀 The video introduces an alternative to the mid-journey AI stable diffusion website, which is not free and has usage limitations based on subscription plans.
  • 💡 The presenter demonstrates how to run mid-journey as a local service on a personal computer, enabling unlimited free image generation.
  • 🎨 The process is based on training data from actual mid-journey images, aiming to produce results similar to the original service.
  • 🛠️ To set up this local service, one must follow a previous tutorial on setting up Automatic 1111 on a Windows machine.
  • 🤖 Hugging Face is highlighted as a valuable resource for various datasets, models, and other tools usable for stable diffusion.
  • 🌐 The 'Open Journey' model, an open-source mid-journey trained model, can be downloaded from Hugging Face and used with Automatic 1111.
  • 🔍 The presenter suggests using prompt sites, such as Creative Indie, for mid-journey style prompts to input into Automatic 1111.
  • 🎭 The results showcased include a variety of styles and themes, such as surreal underwater worlds, historical battle scenes, and steampunk airships.
  • 🔧 The presenter details the settings and parameters used for the image generation, including sampling method, steps, and upscalers.
  • 📈 Prompt Hero is recommended as a website that provides a selection of the best open Journey prompts and examples of generated images with their parameters.
  • 🌟 The video emphasizes the potential of generating highly detailed and specific style images through refined prompts and experimentation with different models.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is about running mid-journey as a local service on one's own PC to generate unlimited free images, based on training data from actual mid-journey images.

  • What is mid-journey and what are its limitations?

    -Mid-journey is an artificial intelligence stable diffusion website that generates artwork. Its limitations include a monthly subscription fee and a limited number of images that can be created based on the subscription plan.

  • What is the significance of the term 'stable diffusion' in the context of the video?

    -In the context of the video, 'stable diffusion' refers to a type of artificial intelligence model used for generating images. It is the underlying technology for platforms like mid-journey.

  • How does one set up mid-journey locally on a PC?

    -To set up mid-journey locally, one needs to follow a tutorial on setting up Automatic 1111 on a local Windows machine. This involves downloading a checkpoint file from Hugging Face and placing it into the models directory of the stable diffusion install.

  • What is Hugging Face and how does it relate to the video content?

    -Hugging Face is a website that hosts various training datasets, models, and other resources for AI applications like stable diffusion. In the video, it is used to find and download an open source mid-journey trained model for local use.

  • What are some features of the open Journey model used in the video?

    -The open Journey model is an open source mid-journey trained model that can generate images similar to those produced by the mid-journey platform. It allows for unlimited free image generation when used locally.

  • How can one find prompts for the stable diffusion model?

    -One can find prompts by visiting websites like Creative Indie and Prompt Hero, which provide a list of mid-journey style prompts and examples of generated images.

  • What are some tips for refining the prompts for better image generation?

    -Tips for refining prompts include using specific artist names in the prompt to stylize the generated images, adjusting the sampling steps for higher quality, and using negative prompts to exclude undesired features.

  • What is the role of the Discord Channel mentioned in the video?

    -The Discord Channel mentioned in the video serves as a community space for users to share their results, discuss their findings, and potentially discover better models or prompts for image generation.

  • What is the significance of the 'high-res fix' and 'upscale' features used in the video?

    -The 'high-res fix' and 'upscale' features are used to increase the resolution of the generated images. 'High-res fix' adjusts the image to a higher resolution, while 'upscale' uses algorithms like ESRGAN 4X to enhance the image quality and detail.

  • What are the potential benefits of using a local service for image generation compared to a subscription-based platform?

    -Using a local service for image generation allows for unlimited free image creation without the constraints of a subscription model. It provides more flexibility and cost-saving for users who require a high volume of image generation.

Outlines

00:00

🚀 Introducing a Free Alternative to Mid-Journey AI Art Generator

The paragraph discusses an alternative method to using the Mid-Journey AI art generation platform by running it as a local service on one's PC. This allows for the creation of unlimited free images based on training data from actual Mid-Journey images. The author explains that this method requires following a previous tutorial on setting up Automatic 1111 and introduces Hugging Face, a website with various models and training datasets for stable diffusion, highlighting the open Journey model which is an open-source Mid-Journey trained model. The simplicity of using this model with Automatic 1111 is emphasized, and the author encourages viewers to explore different models and share their results.

05:01

🎨 Exploring Prompts and Styles for AI-Generated Art

This paragraph delves into the use of different prompts and styles for generating AI art, demonstrating how selecting an artist in the prompt can stylize the generated images. The author showcases various examples, including a historical epic battle scene, a steampunk airship, and a detailed robot, emphasizing the intricate details and epic nature of the results. The use of prompt-building websites like Prompt Hero is introduced to help users refine their prompts and achieve specific styles with less effort. The author also shares their experiences with different prompts and the resulting images, highlighting the potential of this free alternative to Mid-Journey for generating high-quality, detailed art.

Mindmap

Keywords

💡Artificial Intelligence

Artificial Intelligence (AI) refers to the simulation of human intelligence in machines that are programmed to think and learn like humans. In the context of the video, AI is used to generate artwork through a platform called 'stable diffusion', which is a technology that creates images based on textual descriptions or prompts.

💡Mid-Journey

Mid-Journey is an artificial intelligence-based platform that allows users to generate artwork by connecting to Discord and providing textual prompts. It is noted for its high-quality outputs but requires a paid subscription and has limitations on the number of images that can be created based on the user's subscription plan.

💡Stable Diffusion

Stable Diffusion is a type of AI model that is used to generate images from textual descriptions. It is an open-source technology that can be utilized locally on a user's PC, allowing for the creation of unlimited images without the need for a subscription.

💡Discord

Discord is a communication platform that is used by mid-Journey to connect with users and facilitate the process of generating artwork. It is a platform where users can interact with the AI and receive their generated images.

💡Open Journey

Open Journey is an open-source version of the mid-Journey model, which is trained to generate images using AI. It is available for users to download and use with stable diffusion, providing a free alternative to the paid mid-Journey service.

💡Hugging Face

Hugging Face is a website that hosts a variety of AI models and datasets, including the Open Journey model. It is a resource for developers and users looking to access and utilize AI technologies for different purposes, such as image generation.

💡Prompts

Prompts are textual descriptions or phrases that guide the AI in generating specific types of images. They are crucial in determining the output of AI image generation models like stable diffusion, as they provide the context and direction for the AI to follow.

💡Neural Network

A neural network is a series of algorithms that attempt to recognize underlying relationships in a set of data through a process that mimics the way the human brain operates. In the context of AI-generated artwork, neural networks are used to interpret prompts and create images that match the given descriptions.

💡Upscale

Upscaling refers to the process of increasing the resolution of an image, typically by using an AI model to add more pixels without losing detail or clarity. In the video, upscalers like ESRGAN 4X are used to enhance the quality of the AI-generated images.

💡Community and Collaboration

Community and collaboration are central themes in the video, emphasizing the sharing of resources, knowledge, and results among users of AI image generation platforms. The video encourages viewers to join a Discord channel and share their generated images, fostering a sense of community and collaborative learning.

💡Prompt Hero

Prompt Hero is a website that provides a collection of prompts and resources for users of AI image generation models like stable diffusion. It helps users find and refine prompts to achieve specific styles and outcomes in their generated images.

Highlights

The introduction of a method to run mid-journey as a local service on a personal computer, enabling the generation of unlimited free images.

The comparison of mid-journey with an AI stable diffusion website, highlighting the limitations and costs associated with the latter.

The mention of a free trial with limited resources and various subscription plans for the AI stable diffusion website, including a Pro Plan offering 30 hours of fast GPU time and 12 concurrent jobs.

The discovery of an open source mid-journey trained model available on hugging face, which can be used with stable diffusion hosted locally.

The simplicity of downloading and using the open Journey model with automatic 1111 for stable diffusion, by just placing the checkpoint file in the models directory.

The availability of numerous data sets and models on hugging face for different purposes, including 21 different data sets for mid-journey alone.

The setup of a Discord Channel for sharing results and finding better models, encouraging community interaction and collaboration.

The use of creativeindie website with 644 mid-journey style prompts, allowing for easy integration with automatic 1111 for generating images.

The demonstration of generating an image with a surreal underwater world prompt in the style of HR Giger, showcasing the level of detail and intricacy possible with the local service.

The explanation of the sampling method, number of sampling steps, and the use of an upscaler to increase image resolution, providing insights into the technical aspects of image generation.

The presentation of a historical epic battle scene in the style of Frank Frazetta, illustrating the ability to generate highly detailed and stylized images.

The exploration of a steampunk airship prompt in the style of Jules Verne, emphasizing the generation of intricate details and the potential for creative applications.

The tip about websites offering various methods for generating prompts, such as prompt hero, which provides a section on the best open Journey prompts.

The demonstration of using a highly detailed robot prompt from prompt hero, showing how to achieve a specific style with stable diffusion.

The attempt at generating a photorealistic image of a young South Korean girl with mixed results, highlighting the potential and challenges of the local mid-journey service.

The conclusion that the local service provides a free alternative to mid-journey, offering significant creative possibilities on a personal computer.