Microsoft’s FREE Bing AI Art Generator vs Midjourney V5 Direct Prompt Comparison

MattVidPro AI
21 Mar 202331:34

TLDRIn this video, the host explores the new Dolly 2 algorithm integrated into Microsoft's Bing AI Art Generator and compares its image generation capabilities with Midjourney V5. The host discusses the improvements made to Dolly 2, noting its increased detail, creativity, and competitiveness with Midjourney. Despite initial difficulties in generating images through Bing's chat feature, the host successfully generates images using the separate website and Microsoft Edge. The comparison reveals that while the new Dolly 2 produces higher quality images than its predecessor, Midjourney V5 still outperforms in terms of coherence and detail. The video also highlights the new features of Bing's image generator, including the ability to create images through chat and the integration of visual stories and knowledge cards. The host concludes that while the new Dolly 2 is a significant upgrade and offers free image generation, it has yet to surpass Midjourney V5 in overall image quality and versatility.

Takeaways

  • 🔍 Microsoft has integrated OpenAI's DALL-E algorithm into Bing as an image generator, accessible through the new Bing chat feature and Microsoft Edge.
  • 🆕 The new DALL-E 2 algorithm, as part of Bing's image Creator, has been updated to produce more detailed and creative images, competitive with Midjourney.
  • 🌐 Initially, Bing's image generation feature is only available in English and directly integrated into Microsoft Edge, marking it as the first browser with an integrated AI-powered image generator.
  • 📈 The updated DALL-E 2 within Bing can generate images at a resolution of 1024 by 1024, offering detailed and realistic results, surpassing the original DALL-E 2 in quality.
  • 🎨 Bing's image generation tool is designed to cater to the growing demand for visual search experiences, offering a more engaging way to search and interact with content.
  • 🐊 In direct comparisons, the new DALL-E 2 outperformed the original version, producing higher quality images with more coherence and detail, especially noticeable in subjects like crocodiles and anthropomorphic lemons.
  • 📈 The new DALL-E 2 also introduced a 'boost' feature, allowing for faster image generation at the cost of additional credits, which can be replenished weekly or redeemed with Microsoft rewards points.
  • 🚀 Midjourney V5 continues to lead in image generation quality, producing highly realistic and detailed images that are often considered superior to both the original and updated DALL-E versions.
  • 🤖 The original DALL-E 2, despite being outperformed by the new version, still delivered some good results, especially in maintaining the realism of certain subjects like cats.
  • 🎭 When tasked with generating complex prompts that combine different elements, such as a 1940s detective frog or a road sign featuring Walter White, the new DALL-E 2 showed significant improvement over the original version, but Midjourney V5 still excelled.
  • 🆓 One of the significant advantages of the new DALL-E 2 is that it is currently available for free, offering users technically infinite image generations, albeit with potential delays for additional 'fast boost' credits.

Q & A

  • What is the new feature Microsoft has decided to incorporate into Bing?

    -Microsoft has decided to incorporate OpenAI's DALL-E algorithm into Bing, specifically through a feature called Bing Creator, which is an image generator.

  • How does the new DALL-E algorithm compare to Midjourney in terms of image generation?

    -The new DALL-E algorithm appears to be competitive with Midjourney, producing detailed, interesting, and creative imagery.

  • What is the significance of the human brain processing visual information faster than text?

    -The human brain processes visual information about sixty thousand times faster than text, which is why Bing is creating visual tools as it's a critical way that people search for information.

  • How can users access the new Bing Creator image generator?

    -Users can access Bing Creator through Microsoft's new Bing chat feature or directly in Microsoft Edge, making it the first and only browser with an integrated AI-powered image generator.

  • What are the limitations of the new DALL-E algorithm when compared to Midjourney V5?

    -While the new DALL-E algorithm has improved significantly, Midjourney V5 still outperforms it in terms of photorealism, coherency, and clear detail in the generated images.

  • What is the pricing model for the new Bing Creator image generator?

    -The new Bing Creator image generator is currently free to use, with users having access to an infinite number of image generations, although the generation process might take longer without using a boost.

  • How does the new DALL-E algorithm handle complex prompts?

    -The new DALL-E algorithm can handle complex prompts, generating detailed and coherent images that align well with the given prompts, showing significant improvement over the previous version.

  • What are the benefits of using the new DALL-E algorithm for content creators?

    -Content creators can benefit from the new DALL-E algorithm by generating unique and detailed images for their content, which can enhance visual storytelling and engage audiences more effectively.

  • How does the new DALL-E algorithm integrate with other Microsoft services?

    -The new DALL-E algorithm is integrated within the Bing chat experience and is also accessible through Microsoft Edge, providing a seamless experience for users across Microsoft's ecosystem.

  • What are the potential use cases for the new Bing Creator image generator in marketing and advertising?

    -The new Bing Creator image generator can be used to create visually appealing advertisements, social media posts, and marketing materials that can capture attention and convey messages effectively.

  • How does the new DALL-E algorithm compare to its predecessor in terms of image quality and generation speed?

    -The new DALL-E algorithm produces higher quality images with more detail and clarity compared to its predecessor. However, it may take longer to generate images due to the increased complexity and quality.

Outlines

00:00

🤖 Microsoft's Integration of Dolly Algorithm into Bing

The video discusses Microsoft's collaboration with OpenAI to incorporate the Dolly algorithm into Bing's search engine. The Dolly algorithm, previously featured in a video, is noted for its detailed and creative image generation capabilities. The host expresses excitement about testing the new Dolly 2 algorithm through Bing's chat feature and the Bing Creator image generator. The script also mentions issues encountered when attempting to generate images through chat, but successfully finds images of cats eating sushi in a cinematic style. The host further explores the capabilities of Bing Image Creator, noting its integration with over a hundred million chats with the Bing AI, which uses gpt4. The segment ends with a demonstration of generating an image of a crocodile using the new Dolly algorithm within Microsoft Edge.

05:02

🖼️ Comparing Dolly 2 and Bing's Image Generation

The host compares the image generation capabilities of Dolly 2 and Bing's new algorithm. After generating crocodile images, the host notes that Bing's images are more detailed and realistic compared to Dolly 2's, which appear blotchy and incoherent. The host also experiments with more complex prompts, such as an anthropomorphic lemon character in a vaporwave style, and finds that Bing's algorithm generates more coherent and detailed images. The segment explores the use of boost credits to speed up image generation and mentions the possibility of redeeming more credits with Microsoft rewards points. The host concludes that while Bing's algorithm shows significant improvement over Dolly 2, it still falls short of the quality produced by Mid Journey V5.

10:02

🐱 Analyzing the Quality of Generated Cat Images

The video script details the host's analysis of generated cat images using different AI models, including the new Bing AI generation (referred to as Dolly too), the older Dolly 2, and Mid Journey V5. The host notes that while the new Bing AI generation has improved, it still struggles with details like whiskers, appearing cobweb-like. In contrast, Mid Journey V5 produces sharper, more realistic images. The host also compares the results of a prompt involving a person walking through a jungle, noting that Mid Journey V5's images are more detailed and maintain the point of view aspect of the prompt better than the new Bing AI or Dolly 2.

15:04

🧩 Testing Character and Concept Generation

The host tests the AI models' ability to generate complex character concepts and combinations of different famous characters. The script describes the generation of a Walter White Lego character set, with Mid Journey V5 producing highly realistic and coherent images that closely match the prompt. The new Dolly algorithm also performs well, offering a significant improvement over the original Dolly 2, but still not reaching the level of Mid Journey V5. The host also evaluates a prompt for a tabby cat, noting that while the new Dolly algorithm's results are more realistic and sharp, they still don't quite match the quality of Mid Journey V5.

20:06

🎨 Evaluating Artistic and Themed Image Generation

The host evaluates the AI models' performance on artistic and themed image generation, such as a 1940s detective frog and a road sign warning of Walter White's presence. Mid Journey V5 excels in these tasks, producing detailed and coherent images that meet the prompt's requirements. The new Dolly algorithm shows improvement over the original Dolly 2, with better detail and coherence, but still falls short of Mid Journey V5's quality. The host also notes that the new Dolly algorithm struggles with generating images of people smoking, a challenge for AI image generation.

25:07

📸 Reviewing Dolly 2's Image Generation for Various Scenarios

The host reviews the new Dolly 2 algorithm's performance on various image generation scenarios, including a professional photo of a Shitsu, a skateboarding penguin with a sun hat, and a low-angle photo of a Shitsu on a pirate ship. The results are compared to Mid Journey V5, which is found to produce nearly flawless images. The new Dolly 2 algorithm provides acceptable results, significantly better than the original Dolly 2, but still not on par with Mid Journey V5. The host also attempts to recreate a logo with the new Dolly 2, resulting in sharp 3D renders with some over-processing on the edges.

30:08

💬 Discussion on Dolly 2's Pricing Model and Viewer Engagement

The host discusses the new Dolly 2 algorithm's pricing model, which offers technically infinite image generations for free but may take longer without using boost credits. The host contrasts this with Mid Journey's subscription plans and acknowledges that while the new Dolly 2 is a significant improvement over its predecessor, it's hard to compete with Mid Journey V5. The host invites viewers to share their thoughts on the new model, its competitiveness with other models, and the new pricing approach. The video concludes with thanks to the viewers for their support and engagement.

Mindmap

Keywords

💡Bing AI Art Generator

The Bing AI Art Generator is a tool developed by Microsoft that utilizes the Dolly algorithm to create images based on textual prompts. It is integrated into the Bing search engine and is showcased as a competitive feature against other image generation platforms like Midjourney. In the video, the host is excited to explore the new capabilities of this tool, particularly after updates that have improved its image generation quality.

💡OpenAI

OpenAI is a company that collaborates closely with Microsoft and is known for its advanced AI algorithms. The Dolly algorithm, which is a focus of the video, is an OpenAI creation that has been incorporated into Bing's services. The video discusses how Microsoft's decision to integrate this algorithm into Bing is a significant development in the field of AI-generated art.

💡Dolly Algorithm

The Dolly algorithm is an AI model developed by OpenAI that is capable of generating detailed and creative images from textual descriptions. In the context of the video, the host discusses the improvements made to the Dolly algorithm and how it has been integrated into Bing as the Bing AI Art Generator, resulting in high-quality image outputs that rival other platforms.

💡Midjourney V5

Midjourney V5 is an advanced AI image generation platform that the host of the video uses as a benchmark for comparing the capabilities of the Bing AI Art Generator. Throughout the video, the host compares the image outputs of Bing's Dolly algorithm with those of Midjourney V5, noting the strengths and weaknesses of each in terms of detail, creativity, and realism.

💡Image Generation

Image generation refers to the process of creating visual content from textual descriptions using AI algorithms. It is the central theme of the video, as the host explores the capabilities of the Bing AI Art Generator and compares them with other platforms. The video provides examples of generated images, such as crocodiles, anthropomorphic lemons, and various other prompts, to demonstrate the effectiveness of different AI models in creating detailed and coherent visuals.

💡AI Powered Visual Stories

AI powered visual stories are narratives that combine both written and visual content, generated by AI models like the Dolly algorithm. The video discusses how Bing is leveraging this technology to provide users with a more engaging and interactive search experience. These visual stories are designed to offer a more dynamic way of presenting information, making it easier for users to consume and understand complex data.

💡Microsoft Edge

Microsoft Edge is a web browser developed by Microsoft that, according to the video, has become the first and only browser with an integrated AI-powered image generator through its Bing Image Creator feature. This integration is presented as a significant advancement, suggesting that Microsoft Edge users can enjoy a unique and enhanced browsing experience with the ability to generate images directly within the browser.

💡Photorealism

Photorealism is the quality of an image appearing extremely realistic, similar to a photograph. The video frequently references photorealism when evaluating the output of the AI-generated images. The host discusses how the new Dolly algorithm and Midjourney V5 both aim to create images that are not only detailed but also closely resemble real-life photographs, which is a key aspect when comparing their performance.

💡GPT4

GPT4 refers to the fourth generation of the Generative Pre-trained Transformer developed by OpenAI. In the video, it is mentioned that the Bing chatbot AI uses GPT4, which is significant because it indicates the level of advancement in natural language processing and understanding that is being utilized to support the chat feature and potentially influence the image generation process within Bing's services.

💡Stable Diffusion

Stable Diffusion is an AI model used for generating images from text descriptions. While not directly the focus of the video, the host makes a speculative reference to it, suggesting that there might be browser plugins that incorporate Stable Diffusion into other browsers like Chrome. This highlights the ongoing development and integration of various AI models in the field of image generation.

💡Microsoft Rewards

Microsoft Rewards is a program where users can earn points by completing certain activities like taking quizzes or participating in daily polls. The video mentions that these points can be redeemed for additional boost credits in the Bing AI Art Generator, which can be used to speed up the image generation process. This indicates an additional incentive for users to engage with Microsoft's services.

Highlights

Microsoft has integrated OpenAI's DALL-E algorithm into Bing through its new Bing Creator image generator.

The new DALL-E algorithm, referred to as Dolly 2, generates highly detailed and creative imagery.

Users can access Bing Creator by using Microsoft's new Bing chat feature.

Bing's image generation is demonstrated through chat, but the user encountered difficulties in getting Bing to create images directly.

Bing's announcement highlights the ability to create images with one's own words through the new Bing chat feature.

The human brain processes visual information much faster than text, which is why Bing is focusing on visual tools.

Bing Image Creator is powered by an advanced version of the DALL-E model, previewed a few weeks prior.

The new Dolly 2 algorithm shows significant progress from its previous version, generating more realistic and high-resolution images.

Bing Image Creator will be fully integrated into the Bing chat experience and is initially rolling out in creative mode.

Microsoft Edge is the first and only browser with an integrated AI-powered image generator.

The new Dolly algorithm is available to generate images directly within Microsoft Edge.

The updated Dolly within Microsoft Edge produced detailed and large images, showing significant improvement over the base Dolly 2.

The new Bing AI image generator allows for the creation of complex prompts, such as an anthropomorphic lemon character in vaporwave style.

Bing Image Creator offers a more engaged way to search and interact with content, providing a visual storytelling experience.

The new Dolly 2 algorithm can generate images more quickly with a boost, which can be replenished weekly or redeemed with Microsoft rewards points.

Comparing the new Bing AI image generator to Midjourney V5, the latter continues to produce higher quality and more photorealistic images.

Midjourney V5 excels in creating images that look more like paintings rather than photos, offering a unique artistic style.

The new Dolly 2 algorithm, despite improvements, still has room for enhancement to compete with Midjourney V5's level of detail and realism.