Midjourney is Free Again... and here's why.

Matt Wolfe
22 Aug 202422:40

TLDRThe AI image generation landscape is rapidly evolving with new models like Idiogram 2.0 offering free access, prompting Midjourney to provide free trials in response. A comparative test of various AI models including Idiogram, Midjourney, Freepick's Mystic, and Leonardo's Phoenix reveals strengths in realism, text incorporation, and prompt adherence. Each model showcases unique capabilities, indicating a competitive market where consumers benefit from diverse, powerful image generation tools.

Takeaways

  • 🌐 AI image generation technology is rapidly advancing with models like DALL-E 2 generating highly realistic and uncensored images, but may face restrictions due to content policies.
  • 🔍 The video discusses the use of the Flux model from Black Forest Labs, which has been featured in recent videos for its ability to create super realistic images and train custom faces.
  • 🆓 Idiogram 2.0, a text-to-image model that integrates text into images, has been released and is available for free to all users, although it lacks some advanced features like control nets.
  • 📈 The presenter tests various AI models by generating images based on four different prompts, focusing on human realism, landscapes, text incorporation, and absurd image creation.
  • 🎨 Idiogram 2.0 produces decent results in all tested categories, but its free tier is limited to 10 credits per day, allowing for about 40 images daily before requiring payment.
  • 🔥 In response to free alternatives like Idiogram and Flux gaining attention, Midjourney has opened its web experience to everyone and offered free trials, allowing users to generate up to 25 images.
  • 🆕 Freepick's Mystic model, a result of the company's acquisition of Magnific, is another new entrant in AI image generation, still in Alpha but showing promising results in the tests.
  • 🎭 Leonardo's Phoenix model is praised for its high-quality image generation, particularly in terms of aesthetics and color contrast, and the presenter's advisory role is disclosed.
  • 📊 A comparison of various models is presented in a Figma board, showing the outputs for the same prompts across different AI image generators, highlighting the strengths of each in different areas.
  • 💰 The video concludes by summarizing the pricing and availability of different AI image generation models, noting that competition is driving more options and better quality for consumers.
  • 🏆 The presenter emphasizes that while all models have their strengths, the choice of which to use depends on the specific needs and preferences of the user, such as prompt adherence or text incorporation.

Q & A

  • What is the main topic discussed in the video transcript?

    -The main topic discussed in the video transcript is the recent advancements and competition in AI image generation, focusing on the release of new models and how they compare in terms of features and capabilities.

  • What is the significance of the new model 'Idiogram 2.0' mentioned in the transcript?

    -Idiogram 2.0 is significant because it is a highly advanced text-to-image model that is now available for free to all users, offering a strong alternative to other AI image generators in the market.

  • How does the transcript describe the capabilities of 'Grock 2' in AI image generation?

    -The transcript describes 'Grock 2' as an AI image generation model that is capable of producing realistic and uncensored images, with the exception of adult content, and is likely to be adopted in various applications.

  • What are some of the features that differentiate Idiogram 2.0 from other AI image generators according to the transcript?

    -Idiogram 2.0 is differentiated by its unique model that is not built on top of other foundations like stable diffusion or flux, and its ability to incorporate text into images effectively, which is one of its standout features.

  • What is the current status of 'Midjourney' in the context of the video transcript?

    -In the context of the transcript, 'Midjourney' is noted to be responding to competition by offering free trials and improving its product to stay appealing to users, amidst the release of free and advanced AI image generation models.

  • What are the limitations of the free version of Idiogram 2.0 as discussed in the transcript?

    -The limitations of the free version of Idiogram 2.0 include a daily limit of 10 credits, which allows for the generation of about 40 images per day, after which users have to pay for additional generation capabilities.

  • How does the transcript evaluate the performance of different AI image generators in creating realistic images?

    -The transcript evaluates the performance of AI image generators by testing them with specific prompts and comparing the realism of the generated images, noting that most models have caught up with each other in terms of realism.

  • What is the transcript's comparison of 'Mid Journey' and 'Idiogram' in terms of incorporating text into images?

    -The transcript compares 'Mid Journey' and 'Idiogram' by testing them with prompts that require text incorporation. It finds that 'Idiogram' excels at this, while 'Mid Journey' struggles and does not adhere well to the text component of the prompts.

  • What is the transcript's opinion on the current state of AI image generation tools in terms of prompt adherence?

    -The transcript suggests that 'Dolly 3' is currently the best at prompt adherence, meaning it includes all elements from the prompt in the generated image, while other models may miss some details.

  • How does the transcript suggest the AI image generation landscape is evolving, and what does this mean for consumers?

    -The transcript suggests that the AI image generation landscape is becoming more competitive, with a variety of free and high-quality options available. This evolution means consumers have access to a wider range of tools capable of generating images from their ideas, indicating a win for users.

Outlines

00:00

🌊 AI Image Generation Surges Forward

The paragraph discusses the recent advancements in AI image generation, particularly mentioning the capabilities of the AI model 'Grock 2' from Black Forest Labs. It highlights the realistic image generation abilities of this model, excluding adult content. The speaker has been actively exploring and using the Flux model from the same lab, showcasing its features in previous videos. The paragraph also introduces 'Idiogram 2.0', a new text-to-image model from Idiogram that excels at incorporating text into images and is now freely available to all users. The speaker plans to test various AI image generation models, including Idiogram 2.0, against each other using different prompts to evaluate their performance in generating human realism, landscapes, text incorporation, and absurd images.

05:00

🔥 Testing New AI Models: Idiogram vs. Mid Journey

The speaker tests Idiogram 2.0 using various prompts to generate images, finding the results impressive, especially in text incorporation. They compare Idiogram's capabilities with Mid Journey, noting that while Idiogram offers 10 free image generations per day, Mid Journey has introduced a free trial allowing users to generate up to 25 images. The speaker also conducts a side-by-side comparison of image outputs from both platforms, finding Mid Journey's performance in text incorporation and prompt adherence to be lacking compared to Idiogram. The paragraph concludes with a mention of other emerging AI image generation models and the competitive landscape, suggesting that Mid Journey might be feeling the pressure from these new entrants.

10:02

🎨 Exploring Free Pick's Mystic Model and Other AI Image Generators

The paragraph delves into Free Pick's 'Mystic' model, which is in its alpha stage and offers early access to the speaker. Despite its limited daily generation capacity, the outputs are of high quality. The speaker also mentions other AI image generation models like Leonardo's Phoenix, which they personally find impressive, and compares it with outputs from other platforms. The paragraph discusses the varying levels of realism, text incorporation, and prompt adherence across different models, with some models like Dolly 3 excelling in prompt adherence and others like Firefly 3 struggling with text incorporation. The speaker plans to conduct more in-depth testing and comparisons in future videos.

15:03

📊 Comparative Analysis of AI Image Generation Models

The speaker provides a detailed comparative analysis of various AI image generation models, focusing on their strengths in different areas such as realism, text incorporation, and prompt adherence. They mention that while all models have improved in realism, some like Flux and Mid Journey stand out. For text incorporation, Idiogram and Phoenix are highlighted as leaders, while Firefly and playground V3 lag behind. The paragraph also touches on the subjective nature of aesthetics, with the speaker expressing a personal preference for the aesthetics of Leonardo's Phoenix. The speaker plans to expand their testing with more prompts and models to provide a comprehensive guide for users looking to choose the best AI image generation tool for their needs.

20:04

💸 Pricing and Availability of AI Image Generation Tools

The final paragraph discusses the pricing and availability of the AI image generation models mentioned throughout the script. It outlines the free options and limitations of each model, such as Idiogram's 10 free images per day and Mid Journey's 25-image free trial. The speaker also mentions that some models like Meta's Emu are free within certain platforms, while others require subscriptions or are still in alpha testing. The paragraph concludes by emphasizing the benefits of competition in the AI image generation space, which offers users a wide range of options to suit their specific needs and preferences. The speaker invites viewers to explore the AI tools they've showcased and encourages feedback and engagement.

Mindmap

Keywords

💡AI image generation

AI image generation refers to the use of artificial intelligence algorithms to create images based on textual descriptions or other input data. It's a rapidly evolving field that has seen significant advancements, as discussed in the video. The script mentions several models like Idiogram 2.0, Midjourney, and Flux, which are all part of this technology wave, highlighting the surge in innovation and the competitive landscape of AI-driven image creation.

💡Idiogram 2.0

Idiogram 2.0 is a specific model for AI image generation developed by Idiogram. It is noted for being advanced in text-to-image modeling and is capable of incorporating text into images effectively. The video script discusses Idiogram 2.0's release and its current free availability, positioning it as a strong competitor in the AI image generation space.

💡Midjourney

Midjourney is another AI image generation model mentioned in the script, which has recently made its web experience open to everyone and offered free trials. The term is used to illustrate the competitive response to other models in the market, suggesting that Midjourney is adapting to stay appealing to users amidst a crowded AI image generation field.

💡Flux

Flux is an AI model discussed in the video that is used for generating realistic images. It is highlighted as being used underneath other models like Grock 2, indicating its foundational role in the AI image generation process. The script also references the creator's use of Flux in their latest videos, showcasing its capabilities in creating super realistic images.

💡Realism

Realism, in the context of AI image generation, refers to the ability of the models to create images that closely resemble real-world objects, scenes, or people. The script evaluates different models based on their realism, using it as a benchmark for the quality of the generated images. For example, the comparison of the elderly fisherman and the Japanese Zen Garden images demonstrates the models' capacity for realistic depictions.

💡Text incorporation

Text incorporation is the process of including textual elements within generated images, which is a feature that sets Idiogram apart from other models, as per the script. The ability to integrate text with images is crucial for certain applications and is used as a criterion for testing the models' capabilities in the video.

💡Control Nets

Control Nets are a feature in some AI image generation models that allow users to guide the generation process by influencing specific aspects of the image. The script mentions that Idiogram 2.0, while powerful, lacks features like Control Nets when compared to other models, which could limit its flexibility in certain use cases.

💡Prompt adherence

Prompt adherence refers to how well an AI model follows the textual description provided by the user to generate an image. The video script uses this term to evaluate models like Dolly 3 and Flux, which are noted for their ability to include multiple elements from the prompt into the generated images accurately.

💡Leonardo Phoenix

Leonardo Phoenix is an AI image generation model praised in the script for its high-quality image outputs and color contrast. The model is also noted for its improvements in text incorporation. As an adviser to Leonardo, the script's author has a personal bias towards it but acknowledges its capabilities in the competitive landscape.

💡Free trial

A free trial, as mentioned in the context of Midjourney and other models, allows users to test the AI image generation capabilities without cost. The script discusses how models like Midjourney are offering free trials to attract users, which is a strategic move in a market with many free or low-cost alternatives.

💡Aesthetics

Aesthetics in the video script pertains to the visual appeal and artistic quality of the generated images. Different AI models are evaluated based on their aesthetics, with some models like Leonardo Phoenix and Midjourney being highlighted for their particularly pleasing color palettes and visual styles.

Highlights

AI image generation has recently seen a surge in innovation with realistic and uncensored images being generated by models like grock 2.

Grock 2 uses the flux one model from Black Forest labs, which has been heavily utilized in recent video demonstrations.

Idiogram 2.0, a new text-to-image model from Idiogram, is now available for free to all users, marking a significant advancement in the field.

Idiogram 2.0 is unique in that it uses its own Foundation models, not built on top of other existing models like stable diffusion or flux.

The video tests Idiogram 2.0 against other AI models in categories such as human realism, landscape, text incorporation, and absurd image generation.

Midjourney has opened up its web experience to everyone and temporarily enabled free trials, possibly in response to the competition.

Midjourney's free trial allows users to generate a total of 25 images before requiring payment.

Freepick's Mystic model, a result of their acquisition of AI upscale platform Magnific, is another new entrant in AI image generation.

Leonardo's Phoenix model is praised for its high-quality image generation and the aesthetic results it produces.

A comparison of various AI models shows that they have mostly converged in terms of realism and quality of output.

Dolly 3 stands out for its prompt adherence, making it ideal for complex image prompts with multiple elements.

Idiogram excels in incorporating text into images, a feature that sets it apart from other models.

Midjourney and Adobe Firefly struggle with text incorporation in images, showing room for improvement in this area.

The video concludes that with the current competition in AI image generation, consumers have a wide array of options catering to different needs.

A Figma board is provided for viewers to compare the quality and details of images generated by different AI models.

The video suggests that as the field of AI image generation becomes more competitive, the quality and accessibility of tools will continue to improve.

A breakdown of the current landscape of AI image generation is provided, highlighting the strengths and weaknesses of various models.

The video encourages viewers to subscribe for more AI tool and news updates, emphasizing the rapidly evolving nature of the field.