Recraft V3: The Mysterious 'Red Panda' AI Image Generator Unveiled

This is the Mysterious Red Panda AI That Dominated Benchmark Leaderboard

Recraft, an artificial intelligence (AI) firm, introduced its latest AI image generation model Recraft V3 last week. Interestingly, this is the same Red Panda AI model that topped the leaderboard of benchmarking platform Artificial Analysis. The company has now formally unveiled the AI model and highlighted its capability. The Recraft V3 is said to offer enhanced capability in text-to-image generation, and can generate long sentences in images without an issue. It is currently available for both free and paid users across desktop and mobile. In a blog post, the AI firm detailed its latest image generation AI model. The company said that the Recraft V3 model was built keeping the needs of professional designers in mind. One of the key highlights of the text-to-image generator is graphic designing. The AI firm claimed that users can specify the exact positions and sizes of text on a design and Recraft V3 can adhere to the prompt. It can also generate complex graphics by combining images and texts to create a range of effects. Another highlight of the AI model is style control. Recraft V3 offers style as an input, allowing users to pick and choose specific styles for images. This means a brand can add its guidelines in the prompt and the image generator will adhere to it in all of its outputs. The company claimed that the model allows for "fine-grained experimentation" as it catches the details of the style. Additionally, the AI model also supports vector image generation, meaning designers can use it to create various vectors and then use it in real-world scenarios by morphing, expanding, or designing on top of it. Further, the company is also offering a suite of AI image editing tools such as AI Eraser, Modify Area, Inpainting, Outpainting, AI Mockuper, Creative and Clarity Upscalers, and more. These tools are aimed at offering increased control over the output. In the blog post, the company highlighted that the AI model was placed on top of the Artificial Analysis benchmark's text-to-image leaderboard with an Elo score of 1172, beating popular AI models by Midjourney, OpenAI, Stability AI, and more. Recraft V3 is currently available to both free and paid users in the desktop app on Canvas, as a mobile app for Android and iOS, as well as via API.

You can now try Red Panda, the image-generating AI that is breaking all records and that no one knew where it came from - Softonic

Its real name is Recraft V3, and it is one of the image generation models with the most potential on the market Red Panda, a mysterious image generator powered by artificial intelligence, surprised the tech world a few days ago by positioning itself as a leader in the Artificial Analysis platform ranking. Now, it is finally known that behind this name is the Recraft V3 model, an advanced tool for designers created by the company Recraft. Design professionals who want to test the capabilities of this model can now do so through a public access platform. Recraft V3 has climbed to the top spot in the text-to-image model rankings of Artificial Analysis, a research platform that allows users to choose between images generated by different models and rate which one they prefer. Thanks to this participatory evaluation, Recraft V3 has proven to be one of the most popular image generators, surpassing renowned competitors like FLUX1.1 from Black Forest Labs and Midjourney. Recraft V3 allows generating images in multiple sizes and with extensive text, something that, according to the company, "distinguishes this model from the rest." Very few AI generators can match this feature, which provides users with great versatility in their projects. The CEO of Recraft, Anna Veronika Dorogush, emphasized to TechRadar the importance of users achieving "the exact result they have in mind," highlighting that Recraft's goal is to give designers greater control over their work. The Red Panda model, or Recraft V3, was tested for four days in the Artificial Analysis "arena" and quickly captured the interest of users, who appreciate its accuracy and flexibility. The platform crowned it as one of the favorites in the world of generative artificial intelligence, consolidating its position in the field of professional design. Currently, Recraft V3 is available on the Recraft website, with different subscription plans ranging from the free option to premium options. Users can access all its features from the website or with the apps for iOS and Android.

The mysterious AI image generator that topped benchmarks is now available - here's how to try it

A mysterious AI image generator known as Red Panda shot up the ranks of the Artificial Analysis Image Arena this week, cementing itself at the top of the charts. While we had no idea who developed Red Panda, or even what, exactly, it was, we now know who makes one of the best AI image generators and where you can try it... enter Recraft V3. Recraft V3 is an AI tool for professional designers and its developer, also called Recraft, has just released its new image generation model 'red_panda'. As we reported yesterday (October 29), the leaderboard king Red Panda was indeed a new model in testing, and it looks like it could be one of the best out there. Recraft V3 secured the number 1 spot on Artificial Analysis' Text-to-Image Model Leaderboard, a crowdsourced research tool that is created by giving users the choice between images created using two different models and asking them which one they prefer. The new model has a lot going for it, including improved image generation compared to Recraft's previous model. Recraft V3 is the first model of its kind to offer image generation of any size and length of text. Recraft says, "This feature sets it apart from others - while very few models can offer similar capabilities, they still can't match Recraft's ability to generate long texts within an image." Anna Veronika Dorogush, Founder and CEO of Recraft says, "We don't just want users to just enter a prompt and see what comes out. Our goal is to help designers realize the exact result they have in mind. Recraft gives designers enhanced control over their work, enabling them to deliver projects that align perfectly with their creative vision. We are committed to making Recraft the essential, daily tool for every designer." Recraft V3 took part in Artificial Analysis' test for four days, and currently has a 1172 ELO - ELO is a skill-based system found in online gaming and chess, which compares performance based on how many times something wins or loses. The image generator has an arena win rate of 72%, meaning users are more often than not picking the new model over its competitors. Recraft V3 topped the leaderboard, beating Black Forest Labs' FLUX1.1 and Midjourney, TechRadar's current pick for the best AI Image generator. The new model is available from Recraft's website and the company offers a range of subscription plans ranging from free to $48 per month (about £37 or AU$73). Users can access all of Recraft V3's capabilities from the website, or via the iOS / Android app.

Mysterious AI Image Generator More Powerful Than Midjourney Breaks Cover

A mysterious AI image generator that beats all other models in benchmark tests has revealed itself. Initially known as "red_panda", it is actually called Recraft. Earlier this week, talk of a powerful AI image generator topping the Artificial Analysis' Text-to-Image Model Leaderboard -- a ranking platform that compares leading text-to-image models based on human preference -- had people wondering who or what was behind the model known only as red_panda. PetaPixel almost reported on it but didn't due to lack of information. Now Recraft AI, a London-based AI design platform headed by Anna Veronika Dorogush, has claimed responsibility for red_panda. "Nice to meet you!" The company writes on X. "We are red_panda, but for friends just Recraft." At the start of the year, it was reported that Recraft had raised $12 million from venture capital investors. "Introducing Recraft V3 -- a revolutionary AI model that thinks in design language. It delivers unprecedented quality in text generation, outperforming models from Midjourney, OpenAI, and others," Recraft writes on X. "It's more than an image generator, it's a powerful tool with enhanced text placement, style control, and the highest quality standards in the industry." Recraft says it has "set a new standard for excellence in image generation" and that it is the only model that can generate long texts on an image. It also stresses that Recraft is built for graphic designers and has the ability for users to "control text size and placement to create detailed, professional-quality visuals" which it says is "perfect for branding, marketing, and complex graphic layouts." It also allows users to use different styles as an input. If a brand or company wants to create a set of homogenous images in a particular style then users can "refine the style candidate until it matches the exact look and feel needed." Recraft V3 has a background removal tool, a Generative Fill-like tool, an outpainting tool, a mockup tool, a vectorize tool, an upscaler tool, as well of course as a comprehensive AI image generator that is available in different styles including "photorealism", "studio photo", "HDR", "Hard flash", and "black and white". In its experiments, PetaPixel found recraft to be reasonably dexterous -- even placing fingers in the right place of an image of a photographer holding a camera. There is a free version where users get 50 free credits daily. Then there are Basic ($10), Advanced ($27), and Pro ($48) packages. You can use Recraft now by visiting its website and there is also an app on iOS and Android.

Watch out, Midjourney -- Recraft just announced new AI image generator model

Recraft, a leading AI graphic design tool -- and the AI company behind the mysterious Red Panda -- has unveiled its latest image generation model called Recraft V3. And it could set a new benchmark for quality among the best AI image generators. With its designer-centric features, this launch highlights Recraft V3's text generation capability within images, allowing for seamless, accurate text of any size and length, a challenge for most existing models. In fact, recent testing on Hugging Face's Text-to-Image Model Leaderboard by Artificial Analysis secures Recraft V3 at the top position with an ELO rating of 1172, surpassing competitors such as Midjourney and OpenAI. With enhanced control over critical design elements, users can now position text, set custom brand colors, and define unique brand styles, allowing Recraft to produce images that align with their creative vision. These latest features directly address designers' pain points by offering an intuitive workflow, giving them control at the forefront. The platform's Infinite Canvas and Collaborative Workflow features further enhance creative freedom, enabling real-time team collaboration and comparison of design iterations. Recraft's founder, Anna Veronika Dorogush, emphasized the model's design-centric approach: "We don't want users to simply enter a prompt and see what comes out. Our goal is to empower designers with control, enabling them to achieve the exact results they envision." Additionally, Recraft has launched an API, providing businesses and developers access to its image generation capabilities. The API supports raster and vector formats, brand color specification, vectorization, upscaling and background removal, making it a more versatile tool for advanced design workflows. Recraft now reaches over 1.5 million users, and over 200 million images have been created with its platform. Recraft's full features can be accessed via the web or mobile app, available on iOS and Android.

This Mysterious AI Image Generator Is Topping Benchmark Leaderboards

Ranking on Artificial Analysis is decided based on crowdsourcing A mysterious artificial intelligence (AI) model is currently ranking on top of a benchmarking platform's leaderboard, but nobody can find any information about it. The mysterious model in question is dubbed Red Panda and it has ranked at the top position on the Artificial Analysis benchmark's text-to-image generation leaderboard. The AI model has outranked major AI image-generating platforms such as Replicate, Midjourney, and Stability.ai. Notably, the ranking on the benchmarking platform is decided using crowdsourced inputs, raising questions about Red Panda's mystery. The mention of the Red Panda AI model was first spotted in an X (formerly known as Twitter) post, which highlighted that the mysterious AI model was seen in the first position of the text-to-image leaderboard. Artificial Analysis uses the Elo system for ranking, which is also used to rank chess players based on their skill level. The process is simple and the platform lets the users decide which models deserve to rank higher compared to the others. It selects two models on a random basis and adds a unique prompt. Based on the image generated, the benchmarking website then lets the users pick which image is a better depiction of the prompt. The method then continues with random models pitted against each other continuously till enough sample size is collected to rank the AI models. Usually, this is a robust method of ranking AI models since a large proportion of the users on the platform are AI enthusiasts and they understand the nuances of what makes an image generator good. However, the Red Panda being at the first rank raises several questions about the ranking system. This is because there is no information about the developers or the company behind the AI model, no details about the architecture or the foundational model, or if it will be launched anytime soon. Red Panda has an Elo score of 1,181 and a median generation time of seven seconds on the leaderboard. It outranks AI models such as Replicate, Ideogram, Midjourney, Stability.ai, Playground AI, DALLE 3, and Amazon Bedrock.

I tested this viral AI image generator, and it does text well - finally! Try it for free

Recraft's newest model - the mystery AI that beat Midjourney and DALL-E - can generate high-quality images with impressive details, quality, and prompt fidelity. How to try it. It's been a while since a new text-to-image generator shook up the generative AI space. However, the mysterious Red Panda generator has done just that, climbing up Artificial Analysis's Text-to-Image Arena leaderboards and beating out leading models. Now, the identity has been revealed. Also: The best AI image generators: Tested and reviewed On Wednesday, Recraft launched its newest model, Recraft V3, the same model that appeared as Red Panda in the Arena. The leaderboard results show that the model can generate high-quality images with impressive details, quality, and prompt fidelity. However, according to Recraft, its standout is its text generation capabilities. Some prompts -- such as those involving hands, faces, and text -- are particularly challenging for image generators. Most attempts at text image generation fall short, getting close but messing up one letter, spelling, or making up random words. However, Recraft claims its model can generate anatomically correct images and accurate long text strings. "The main advantages of Recraft V3 [lie] in text generation quality, anatomical accuracy, prompt understanding, and high aesthetic quality," said Recraft in a blog post. "Recraft V3 is the only model in the world that can generate images with long texts, as opposed to just one or a couple of words." To see if the claims hold up, you can test Recraft V3 for yourself using the instructions below -- or scroll down to see how it fared on my tests. The model is available for free and paid users online and in the mobile app. Getting started is easy: All you have to do is visit the website, click on "Generate AI image," and create a Recraft account or sign in with an existing Google, Discord, Apple, or single sign-on. Also: Can Apple's new AI photo 'Clean up' tool beat Google's Magic Editor? For some users, yes Different plans are available to better suit users' needs, starting with a free plan that offers 50 free credits daily and makes all generated images public. The more advanced plans offer higher limits and more advanced features and range from $10 per month to $48 per month. Once you're in, click on "Create new image," type in a prompt, personalize the settings, and click on "Recraft." It took 15 seconds to generate two images. I tested for quality for the first generation using the prompt, "A vibrant, realistic hummingbird perched on a tree." The results were very impressive and comparable with some of the best image generators' takes on the prompt, which you can see on this list. I included one image below. For the next prompt, I went for something more challenging - hands. I entered the prompt, "Two manicured hands typing on a laptop." The images look OK at first glance. However, when I take a closer look, I can spot some inconsistencies. Lastly, for the most exciting prompt and biggest challenge, I asked it to generate an image of a computer screen that read ZDNET's brand's mission statement, "ZDNET, tomorrow belongs to those who embrace it today," in electric yellow. I included both results for this one because they were equally impressive. Not only was all of the text accurately spelled and transferred, but it was also uniformly displayed and spaced out as if a human had placed it there. It was also layered very nicely onto the backdrop, resulting in realistic photos that look like they were taken by a camera. If you look closely, there is some variation in the uppercased words, but that is minimal compared to text results from most other generators that can't even get the letters out.

New mystery AI image generator bests Midjourney and DALL-E 3 - how to try it

Red Panda may be the new AI image generator to beat, yet its origins are unknown. Even though many AI image generators are vying for your attention today, they each have different strengths and, as a result, different use cases. However, all AI can be judged on certain metrics -- such as the overall quality of the rendition -- and that's what Artificial Analysis's Text-to-Image Arena tests are for. Also: The best AI image generators: Tested and reviewed In the Arena, users are shown two images rendered from the same prompt without labeling which model is responsible for each. Then, users can pick which they prefer based on overall quality and factors such as lack of artifacts and prompt fidelity. The results are compiled into a leaderboard with a new mystery model, Red Panda, at the top of the charts. The new model beats out many industry leaders featured in the Arena, including DALLE 3, Stable Diffusion 3 Large, Midjourney v6, and FLUX.1 [pro]. The origins of Red Panda are unknown, but it is likely a new model that the company behind it is testing in the Arena to gauge performance before launching to the public, a common industry practice. Red Panda has a 72% Arena Win Rate and an Arena Elo score of 1172, which puts it 40 points ahead of its runner-up, FLUX 1.1 [pro] by Black Forest Labs, the model behind the Grok image generator found on X. To experience the Red Panda hype for yourself, visit the Text-to-Image Arena and run through the selection process. When you enter the Arena, all you have to do is select which photo matches the prompt better; once you do, you will see which models were used and which you voted for. However, there is no guarantee how soon you will be presented with Red Panda; you may have to go through several rounds until it comes up. It came up several times within my first 30 rounds, which sounds like a lot, but it takes only seconds to go through each pick. Then, you can view your voting results by visiting your personal leaderboard; you can view mine below. The results of my 60 rounds confirmed Red Panda is an extremely capable model. It won four times, tying with Midjourney --which ZDNET crowned the best AI image generator for high-quality photos -- and FLUX.1. Also: Can't find Apple's AI features after upgrading to iOS 18.1? Do this Stay tuned for ZDNET coverage when more information about the model is released. Until then, there are plenty of other AI image generators you can test out, and ZDNET rounded up the best ones for you here.

Midjourney has a new challenger for the best AI image generator but nobody has heard of it before

Midjourney has a new challenger for the title of best AI image generator, and it's a model nobody has heard of before. The AI image generator in question is called "red_panda" and is currently top of the crowdsourced Artificial Analysis rankings for text-to-image generation. Artificial Analysis' arena leaderboard is created by giving users the choice between two images and asking them which one they prefer. It's all down to a matter of taste but as it stands red_panda has a 79% win rate beating out rivals like Black Forest Labs, Ideogram, and Midjourney. So what is red_panda you ask? Well, no one actually knows but Artificial Analysis can be a great way to test AI models before launching to the general public and this may be the case here. The name of the AI could give us a hint, however, after all, red pandas are found in the eastern Himalayas and southwestern China. Could this AI image generation model be a new contender from a large Chinese company? As it stands, red_panda has a 1244 ELO and has been selected 1492 times, which is way less than the next best AI model, Black Forest Labs' FLUX1.1 Pro. ELO is a skill-based system found in online gaming and chess, which compares performance based on how many times something wins or loses. With a 79% win rate, it's fair to say that red_panda is definitely appealing to the eye. Red_panda might be a new Chinese AI model, or it could be the next evolution in a major player like Midjourney currently in development before its official reveal. One thing is for sure, whatever red_panda is we should know soon as topping any leaderboard is great marketing for the launch of a new product. As it stands, Midjourney is TechRadar's pick for the best AI image generator but we'll be sure to test red_panda out as soon as it becomes available to see if it truly is as good as Artificial Analysis' users think it is.

A new AI image generator has popped up without a named creator or any explanation, and it's seemingly already surpassing its competition

I'm already quite sceptical of AI image generation. I'm cautious of the way it seems to flippantly skirt around (or entirely ignore) copyright, and I worry about the negative feedback loop of replacing artists from making art. Caution is important to practice, especially now that red_panda, a brand-new image-generation tool, has just been spotted on a site dedicated to voting for the best AI and rocked straight up to the top of the leaderboards. As reported by TechCrunch, a site called Artificial Analysis has a section called "Arena" that allows users to vote between two images on which fits a given prompt best -- the two images being generated by different AI image generation models. In the leaderboard, red_panda sits at the very top, with a 72% win rate. This means that in all the battles its images have found themself, it has only lost 28% of them. Some in the comments of Artificial Analysis' post claim this could be a new version of Midjourney or Baidu's AI tool, and this could certainly be the case given the creator's name is obscured, but only time will tell what exactly red_panda is. This new tool has the highest ELO (generally perceived as a metric of quality) in the data pool and a fast generation time at seven seconds, but no price per image, as it's all fairly unknown. The win rate is and impressive figure, and that is why it was noted by the Artificial Analysis X account. At the time, it had an even more impressive track record, at a 79% win rate. The red panda name does perhaps feel in poor taste with the energy costs of AI models like this, given it is an endangered animal partially down to deforestation and climate change. The images given in the battle aren't created in the moment, based on generation time, as some models can take up to a minute to generate an image. If you aren't a fan of AI image generation, playing this game doesn't seem to directly contribute to it in a meaningful way. I tried voting in the "Arena" myself and I came away quite surprised by my own personal leaderboards. Within 60 choices, red_panda had only come up two times and sat at the very top of my leaderboard, with a 100% win rate. By 150, that win rate dropped all the way down to 61% for me with 11 total selections. Two different versions of the FLUX model hogged the top of my personal leaderboard. These same two models sit at two and three in the global leaderboard. The AI model red_panda mostly took those wins early on from me because it's one of a handful of models that can actually handle text with any competency. It is also pretty good at removing that ill-proportioned glossiness that many AI images tend to have. The site itself regularly ran into problems, once running the same image from the same model against itself. Then, one time, I was given three images (one of which was the wrong prompt and couldn't be clicked) and the other two had all of their text wrong. Another win red_panda managed to win was against an image of colourful static, where the model or site simply failed to display correctly. It's also worth noting I ran into duplicate images in that time so these figures could technically be weighted, if a company wanted that to happen. Though this image generation tool does seem pretty impressive, based on selection figures, it's important to note that much of its competition is pretty poor, its win rate has gone down with time, and it has been selected a total of just under 15,000 times in battles. We don't yet know what it is, how it works, or where it scrapes the imagery it is trained on. From my brief time testing it, the model seems to be on the higher end of AI imagery generation tools but I still spotted some problems in the arena.

A mysterious new image generation model has appeared | TechCrunch

A mysterious new image generation model is beating models from Midjourney, Black Forest Labs, and OpenAI on the crowdsourced Artificial Analysis benchmark. The model, which goes by the name "red_panda," is around 40 Elo points ahead of the next-best-ranking model, Black Forest Labs' Flux1.1 Pro, on Artificial Analysis' text-to-image leaderboard. Artificial Analysis uses Elo, a ranking system originally developed to calculate the relative skill level of chess players, to compare the performance of the various models it tests. Similar to the community AI benchmark Chatbot Arena, Artificial Analysis ranks models through crowdsourcing. For image models, Artificial Analysis selects two models at random and feeds them a unique prompt. Then, it presents the prompt and resulting images, and users choose which they think better reflects the prompt. Granted, there's some bias in this voting process. Artificial Analysis' voters are AI enthusiasts, for the most part, and their choices might not reflect the preferences of the wider community of generative AI users. But red_panda is also one of the better-performing models on the leaderboard in terms of its generation speed. The model takes a median of around 7 seconds to generate an image -- over 100 times faster than OpenAI's DALL-E 3. So, where did red_panda come from? Which company made it? And when can we expect it be released? All good questions. AI labs increasingly use community benchmarks to drum up anticipation ahead of an announcement, though, so it might not be long before we find out.

I tested this viral AI image generator and it really can do hands, faces, and text - for free

Recraft's newest model - the mystery AI that beat Midjourney and DALL-E - can generate high-quality images with impressive details, quality, and prompt fidelity. How to try it. It's been a while since a new text-to-image generator shook up the generative AI space. However, the mysterious Red Panda generator has done just that, climbing up Artificial Analysis's Text-to-Image Arena leaderboards and beating out leading models. Now, the identity has been revealed. Also: The best AI image generators: Tested and reviewed On Wednesday, Recraft launched its newest model, Recraft V3, the same model that appeared as Red Panda in the Arena. The leaderboard results show that the model can generate high-quality images with impressive details, quality, and prompt fidelity. However, according to Recraft, its standout is its text generation capabilities. Some prompts -- such as hands, faces, and text -- are particularly challenging for image generators. Most attempts at text image generation fall short, getting close but messing up one letter, spelling, or making up random words. However, Recraft claims its model can generate anatomically correct images and accurate long text strings. "The main advantages of Recraft V3 [lie] in text generation quality, anatomical accuracy, prompt understanding, and high aesthetic quality," said Recraft in a blog post. "Recraft V3 is the only model in the world that can generate images with long texts, as opposed to just one or a couple of words." To see if the claims hold up, you can test Recraft V3 for yourself using the instructions below -- or scroll down to see how it fared on my tests. The model is available for free and paid users online and in the mobile app. Getting started is easy: All you have to do is visit the website, click on "Generate AI image," and create a Recraft account or sign in with an existing Google, Discord, Apple, or single sign-on. Also: Can Apple's new AI photo 'Clean up' tool beat Google's Magic Editor? For some users, yes Different plans are available to better suit users' needs, starting with a free plan that offers 50 free credits daily and makes all generated images public. The more advanced plans offer higher limits and more advanced features and range from $10 per month to $48 per month. Once you're in, click on "Create new image," type in a prompt, personalize the settings, and click on "Recraft." It took 15 seconds to generate two images. I tested for quality for the first generation using the prompt, "A vibrant, realistic hummingbird perched on a tree." The results were very impressive and comparable with some of the best image generators' takes on the prompt, which you can see on this list. I included one image below. For the next prompt, I went for something more challenging - hands. I entered the prompt, "Two manicured hands typing on a laptop." The images look OK at first glance. However, when I take a closer look, I can spot some inconsistencies. Lastly, for the most exciting prompt and biggest challenge, I asked it to generate an image of a computer screen that read ZDNET's brand's mission statement, "ZDNET, tomorrow belongs to those who embrace it today," in electric yellow. I included both results for this one because they were equally impressive. Not only was all of the text accurately spelled and transferred, but it was also uniformly displayed and spaced out as if a human had placed it there. It was also layered very nicely onto the backdrop, resulting in realistic photos that look like they were taken by a camera. If you look closely, there is some variation in the uppercased words, but that is minimal compared to text results from most other generators that can't even get the letters out.

Mysterious new 'Red Panda' AI image generator appears out of nowhere -- is this a new DALL-E?

So another day, and yet another mysterious AI model screams up the leaderboards in a flurry of activity. The leaderboard in question is the Artificial Analysis Image Arena, a resource commonly used to evaluate the quality of AI image models when they reach the market. The idea is simple. Model owners submit their tech for trial by a public jury. Images are generated through text prompts, and the image results of two different models are displayed side by side on the arena page for public voting. These arenas typically deliver expected outcomes from mainstream players like Flux, Midjourney, Ideogram and StableDiffusion. But occasionally a new model breaks through, like this Red_Panda, and destroys the competition. At this point, rampant social media speculation begins until the identity of the new entrant is revealed. Right now there's a fevered discussion going on about whether the panda is the latest Midjourney variant, something from OpenAI or a complete unknown. The images are anonymous, so no one knows what model they're voting for, it's just a matter of personal taste. The totals are tallied on an ongoing basis, with the leaderboard continually updated from the results. It's a great way for companies to test new models before releasing them to the public, and it is, of course, great for launch marketing if you've come up with something special. The last speculation was around a model called Blueberry that turned out to be Flux 1.1 Pro from Black Forest Labs. This time around my money is on a new model from the Far East. The big Chinese players like Baidu and Tencent have yet to stamp their mark on the image generation sector. It also makes sense if we remember that there are some stunning Chinese AI video generation platforms currently dominating the market -- like Kling and MiniMax, who are slugging it out for the top of the tree. Adding a world-leading AI image model would fit in with the China's clear intent to keep up with the leading companies from the West, and there's no reason why it shouldn't work. I did my test on the Arena and the Red_Panda model easily came top. And bear in mind I am probably a pretty generic user. Right now I use Flux models because they're accurate, free and do text really well. I've also rated Ideogram highly in the past. The last model to burst onto the scene like this was Flux 1.1 Pro, and what a beauty that turned out to be. So perhaps we're about to see a step-change uplift in the quality of our AI art. At that point the other important fact will be if it's a closed or open source model. Never a dull moment in this game.