AI images just got WAY too real. FLUX 1.1 deep dive

AI Search
4 Oct 202433:15

TLDRThe video explores FLUX 1.1, a cutting-edge AI image generator that produces hyper-realistic images. It compares FLUX 1.1 Pro with other leading generators using various prompts, highlighting its strengths in context understanding and realism. The script also reveals techniques like using file names as prompts to generate images that mimic amateur photography, blurring the line between AI and reality. Despite minor flaws in detail, FLUX 1.1 Pro emerges as a powerful tool, set to evolve further with upcoming versions promising even higher resolution and faster generation at a reduced cost.

Takeaways

  • 😲 AI-generated images are becoming incredibly realistic with the release of FLUX 1.1, the latest AI image generator.
  • 🔍 FLUX 1.1 Pro outperforms other leading image generators based on benchmark scores and various metrics.
  • 💰 Users can try FLUX 1.1 Pro on platforms like Together AI and Rubber Band with free sign-ups and credits.
  • 📷 A simple trick to generate more realistic, non-AI looking images is by using prompts like 'IMG uncore' followed by numbers and 'CR2'.
  • 🐱 Adding simple prompts like 'cat' or 'selfie' to the 'IMG uncore' trick can produce convincingly normal, non-AI photos.
  • 📈 FLUX 1.1 Pro is capable of creating images that are nearly indistinguishable from real photos, blurring the line between AI and reality.
  • 🚫 The 'IMG uncore' trick works best with simple prompts; more detailed prompts can reduce its effectiveness.
  • 🤳 Mirror selfies and other specific photo styles can be generated using certain keywords, making FLUX 1.1 Pro highly versatile.
  • 🎭 FLUX 1.1 Pro can also generate images in different styles, such as watercolor paintings, showcasing its adaptability.
  • 🌐 FLUX 1.1 Pro is more affordable than its predecessor, with a lower cost per image, and offers faster generation times and higher resolution capabilities.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is a deep dive into FLUX 1.1, an AI image generator that has been recently released and is considered one of the best in its category.

  • What are some of the platforms where FLUX 1.1 Pro can be used?

    -FLUX 1.1 Pro can be used on platforms such as together.ai, rubberband.com, everart.ai, foul.a, replicates, and free.com.

  • How does the user generate images using FLUX 1.1 Pro on together AI?

    -On together AI, after signing up for a free account, the user gets $5 in credits to generate images with FLUX 1.1 Pro. The user can adjust the width and height and enter a positive prompt to generate images.

  • What is a trick mentioned in the video to generate images that look like they were taken by a regular person?

    -A trick mentioned is to enter 'IMG uncore' followed by any four numbers and 'CR2' as the prompt, which simulates a file structure from a Canon camera, resulting in images that look like amateur photos.

  • How does the video demonstrate the capabilities of FLUX 1.1 Pro in generating realistic images?

    -The video demonstrates the capabilities of FLUX 1.1 Pro by generating images with simple prompts like 'cat', 'selfie', and 'party', and showing how these images closely resemble real, non-AI generated photos.

  • What are some of the file formats and keywords used to simulate different types of photos in the video?

    -The video uses file formats like CR2 and HEIC, and keywords such as 'selfie', 'dinner', 'passport photo', and 'yearbook photo' to simulate different types of photos.

  • What is the significance of the 'CR2' and 'JPEG' prompts in generating images with FLUX 1.1 Pro?

    -The 'CR2' and 'JPEG' prompts are significant because they allow the AI to generate images that mimic the look of photos taken by specific types of cameras or in certain file formats, resulting in more realistic and less AI-like images.

  • How does the video compare FLUX 1.1 Pro with other image generators?

    -The video compares FLUX 1.1 Pro with other leading image generators like FLUX 1 Pro, Ideogram version 2, and Mid Journey version 6.1 by testing them with challenging prompts and evaluating the realism and adherence to the prompts.

  • What are some of the challenges FLUX 1.1 Pro faces according to the video?

    -Some challenges FLUX 1.1 Pro faces include generating images with correct hand and finger details, understanding uncommon animals, and producing text on neon signs or in cityscapes.

  • What are the improvements and new features of FLUX 1.1 Pro over the previous version?

    -FLUX 1.1 Pro offers six times faster generation than FLUX 1 Pro, improved image quality, better prompt adherence, and diversity. It also has the capability to generate ultra-high-resolution images up to 2K and costs less at 4 cents per image compared to the previous version's 5 cents.

Outlines

00:00

🖼️ Introduction to Flux 1.1 Pro AI Image Generator

The paragraph introduces Flux 1.1 Pro, the latest and best AI image generator available. It mentions that the video will compare Flux 1.1 Pro with other leading image generators based on benchmark scores, but the presenter prefers to focus on practical testing. Various platforms where Flux 1.1 Pro can be used are listed, including Together AI, Rubber Band, Ever Art, Foul, Replicates, and Free.com. The presenter chooses Together AI for demonstration due to the free $5 credits, allowing for several dozen image generations. The process is simple: users can input a positive prompt and adjust image dimensions, but there are no advanced settings like CFG scale or negative prompts. The video demonstrates the generation of images starting with a simple prompt like 'a woman in the city' and discusses the typical AI-generated look with polished faces and blurry backgrounds. A trick to generate more realistic images is introduced: using the prompt 'IMG uncore' followed by any four numbers and 'CR2', which simulates a file structure from a Canon camera, resulting in images that look like they were taken by a regular person rather than AI.

05:01

📸 Exploring Prompt Techniques for Realistic AI Images

This paragraph delves into the techniques for generating more realistic images using Flux 1.1 Pro. The presenter demonstrates how changing the prompt to include 'CR2' and numbers can produce images that mimic amateur photography, with grainy textures and non-blurry backgrounds. Examples include generating images with prompts like 'cat', 'selfie', and 'party', which result in photos that closely resemble real, non-AI generated images. The paragraph also highlights the limitations of these techniques, noting that they work best with simple prompts and can fail with more detailed descriptions. The presenter experiments with different file formats like 'heic' and different numbers to see variations in the realism of the generated images. The video also shows how prompts like 'selfie.JPEG' can produce highly realistic selfie photos, and the presenter encourages viewers to be creative with their prompts, suggesting that Flux 1.1 Pro can generate images that are nearly indistinguishable from real photos.

10:01

🕹️ Testing Flux 1.1 Pro with Challenging Prompts

The paragraph discusses the testing of Flux 1.1 Pro with more complex and challenging prompts compared to other top image generators like Ideogram version 2 and Mid Journey version 6.1. The presenter sets the image size to 1024x1024 and uses a series of difficult prompts to test the generators. The first prompt is for a woman doing a Warrior 1 yoga pose at home, which none of the generators handle perfectly, but Google's image gen 3 and Dolly 3 perform slightly better. The presenter notes that Flux 1.1 Pro and Flux 1 Pro produce very similar results, and Mid Journey fails to generate the correct hand and finger details. The second prompt is for a man giving a TED talk with a neon sign and specific lighting details. Flux 1.1 Pro understands the prompt well, generating the TEDEX logo and text correctly, although with some minor flaws. The comparison shows that Flux 1.1 Pro and Flux 1 Pro are quite similar, with minor improvements in 1.1 Pro.

15:02

📱 Comparing AI Image Generators with Realistic Prompts

This paragraph compares Flux 1.1 Pro with other image generators using prompts that aim to produce realistic, low-quality images. The presenter tests the generators with prompts like 'lowquality Snapchat photo of a Teenage man taking a mirror selfie' and 'a teenage woman holding a handwritten note that says verify me 8 2022 2024 lowquality selfie photo'. The results show that Flux 1.1 Pro and Flux 1 Pro produce images that are too polished and blur the background too much for the desired low-quality effect. Ideogram performs better in these tests, producing images that look more like real, low-quality selfies. Mid Journey struggles with text generation and overall realism. The paragraph also includes a sponsorship mention for Turbo Type, a tool for creating custom keyboard shortcuts to save time on repetitive tasks.

20:03

🤚 Testing Flux 1.1 Pro's Ability to Generate Detailed Images

The paragraph focuses on testing Flux 1.1 Pro's ability to generate detailed images with prompts like 'two hands making a heart symbol' and 'a woman showing her palms and soles of feet'. The results for the heart symbol prompt are good, with Flux 1.1 Pro producing a realistic image, although the background is too blurry for an amateur photo. When testing the prompt for a woman showing her palms and soles of feet, Flux 1.1 Pro fails to meet the prompt's requirements, producing an image that does not show the requested details. The comparison with other generators shows that Ideogram comes closest to the prompt, while Mid Journey misses key elements entirely. The paragraph also notes that Flux 1.1 Pro's images tend to be slightly brighter and more saturated compared to other generators.

25:05

🎨 Evaluating Flux 1.1 Pro's Artistic and Stylistic Capabilities

This paragraph evaluates Flux 1.1 Pro's ability to generate images in different styles, starting with a prompt for a watercolor painting of a whale in the sky. Flux 1.1 Pro successfully creates a watercolor-like image that follows the prompt closely. When compared to other generators, Flux 1.1 Pro is preferred for its watercolor effect and accurate depiction of the whale. The paragraph then tests Flux 1.1 Pro's ability to generate anime-style images with the prompt 'an anime girl in the city at night'. While the result is good, the text on the signs in the background is not realistic, which is a common issue with all current image generators. Comparing the generators, Ideogram produces a more anime-style background, while Flux 1.1 Pro and Flux 1 Pro generate images that look like anime illustrations or posters. Mid Journey's result is considered the least anime-like.

30:07

🐉 Testing Flux 1.1 Pro with Uncommon Animals and Styles

The paragraph tests Flux 1.1 Pro's ability to generate images of uncommon animals and different styles. Using the prompt 'a kodo dragon in the forest', Flux 1.1 Pro fails to generate a realistic kodo dragon, which is consistent with its known weakness in generating uncommon animals. Mid Journey outperforms the other generators in this case, producing the only recognizable kodo dragon, although with some inaccuracies. The presenter also experiments with the prompt 'komodo dragon' followed by a random date in JPEG format, but the result is still not a realistic kodo dragon. The paragraph concludes with a summary of the test and comparison of the four best image generators, noting that Flux 1.1 Pro does not show a significant improvement over Flux 1 Pro and shares similarities in performance with its predecessor.

🏆 Flux 1.1 Pro's Performance and Updates

This paragraph discusses the performance of Flux 1.1 Pro based on a third-party evaluation and its updates. Flux 1.1 Pro has the highest Elo score and win rate of 69% according to the image generator leaderboard by Artificial Analysis. It is revealed that Flux 1.1 Pro, previously known as the mysterious 'blueberry' model, outperformed all other models. The paragraph highlights that Flux 1.1 Pro generates images six times faster than Flux 1 Pro while improving image quality, prompt adherence, and diversity. Flux 1 Pro has also been updated to generate images twice as fast as before. Flux 1.1 Pro will soon be able to generate ultra-high-resolution images up to 2K without compromising prompt following, and its pricing is cheaper than Flux 1 Pro at 4 cents per image. The paragraph concludes by encouraging viewers to try the CR2 or JPEG prompts for generating realistic, amateur-looking photos and to share their experiences and prompting techniques.

Mindmap

Keywords

💡AI image generator

An AI image generator is a software tool that uses artificial intelligence to create images based on textual descriptions or prompts. In the context of the video, the AI image generator FLUX 1.1 is discussed as the latest and best in its category, capable of generating highly realistic images. The video script mentions various platforms where users can access and utilize FLUX 1.1 to generate images, emphasizing its capabilities and the quality of its outputs.

💡FLUX 1.1 Pro

FLUX 1.1 Pro is the upgraded version of the AI image generator FLUX, mentioned in the video as providing superior image generation capabilities. It is highlighted for its faster generation speed, improved image quality, and better adherence to prompts compared to its predecessor. The video script compares FLUX 1.1 Pro with other leading image generators, showcasing its performance in generating images from various prompts.

💡Benchmark scores

Benchmark scores are a measure of performance used to compare different systems or models, such as AI image generators. In the video, the speaker refers to benchmark scores to compare FLUX 1.1 Pro with other image generators based on various metrics. These scores help determine which image generator performs best in different aspects, such as realism and prompt adherence.

💡Platforms

Platforms, in the context of the video, refer to the online services or websites that host the AI image generator FLUX 1.1 Pro, allowing users to access and use it. Examples mentioned include Together AI, Rubber Band, Ever Art, and others. These platforms offer different features, such as free credits or images, and some may require payment for more extensive use.

💡Prompts

Prompts are the textual descriptions or commands that users input into an AI image generator to guide the creation of an image. The video discusses how different prompts can yield varying results, from highly realistic images to those that look like they were taken by an amateur photographer. The effectiveness of prompts is demonstrated through various examples, such as 'IMG uncore' and 'CR2', which simulate the file structure of a Canon camera to generate more realistic images.

💡Realism

Realism, in the context of AI-generated images, refers to how closely the generated images resemble real-life photographs or scenes. The video emphasizes the high level of realism achieved by FLUX 1.1 Pro, noting that some images are becoming indistinguishable from actual photos. This is demonstrated through comparisons and discussions of image quality, with the video showing examples of images that look like they were taken by regular people with their phones.

💡Resolution

Resolution in the context of image generation refers to the number of pixels used to form the image, which affects its clarity and detail. The video mentions that FLUX 1.1 Pro will soon be able to generate ultra-high-resolution images, up to 2K, without sacrificing the quality of prompt adherence. This indicates an advancement in the capabilities of AI image generators to produce more detailed and clear images.

💡Pricing

Pricing in the video refers to the cost associated with using the AI image generator FLUX 1.1 Pro. It is highlighted that FLUX 1.1 Pro is more cost-effective than its predecessor, costing 4 cents per image compared to 5 cents for FLUX 1 Pro. This makes FLUX 1.1 Pro not only a better performer but also a more economical choice for users.

💡Closed source

Closed source indicates that the software's source code is not available to the public, meaning users cannot access or modify it. The video mentions that the Pro version of FLUX is closed source, so users cannot download and run it locally; instead, they must use it through the provided platforms or opt for the Dev or Schnell versions for local use.

💡Prompting techniques

Prompting techniques refer to the methods or strategies used to input prompts into an AI image generator to achieve desired outcomes. The video introduces several techniques, such as using file format模拟(e.g., 'CR2' or 'JPEG') to trick the AI into generating images that mimic the style of photos taken from specific devices. These techniques are showcased as ways to enhance the realism and diversity of the images produced by FLUX 1.1 Pro.

Highlights

Flux 1.1 Pro has been released, touted as the best AI image generator currently available.

Benchmark scores and metrics comparing Flux 1.1 Pro to other leading image generators are available for review.

Platforms like Together AI and Rubber Band offer Flux 1.1 Pro for free or with credits upon sign-up.

Ever Art and Foul allow the use of Flux 1.1 Pro, but require payment for credits.

Replicates and Free.com also support Flux 1.1 Pro, with costs around 4 cents per megapixel or per image.

Together AI is used in the video for its $5 credit offer, facilitating dozens of image generations.

Flux 1.1 Pro generates images with typical AI characteristics, such as polished faces and blurry backgrounds.

Using specific file format prompts like 'IMG uncore' and 'CR2' can simulate the look of photos from Canon cameras.

Prompts with 'CR2' or 'heic' followed by numbers can produce images resembling amateur photography.

Adding simple prompts like 'cat' or 'selfie' to the file format trick can generate remarkably normal-looking photos.

Flux 1.1 Pro can create images that are nearly indistinguishable from real photos, blurring the line between AI and reality.

The trick of using file format prompts works best with simple keywords and short prompts.

More detailed prompts can override the effects of the file format trick, returning a typical AI-generated look.

Flux 1.1 Pro struggles with more complex prompts, such as 'woman at a cafe', reverting to a polished, AI appearance.

Using 'selfie.JPEG' can generate grainy, non-blurry images that resemble real selfie photos.

Flux 1.1 Pro can generate highly realistic images, such as mirror selfies and low-quality phone photos.

The video demonstrates Flux 1.1 Pro's ability to generate images that are often mistaken for real photos.

Flux 1.1 Pro's performance is compared to other top models like Flux 1 Pro, Ideogram v2, and Mid Journey v6.1.

Flux 1.1 Pro provides six times faster generation than Flux 1 Pro while improving image quality and prompt adherence.

Flux 1.1 Pro will soon be able to generate ultra-high-resolution images up to 2K.

Pricing for Flux 1.1 Pro is cheaper than Flux 1 Pro, at 4 cents per image.

Flux 1.1 Pro is confirmed to be the mysterious 'blueberry' model that outperformed all other models on the leaderboard.

Flux 1.1 Pro is recommended for its ability to generate realistic, amateur-looking photos that are hard to distinguish from real ones.