World's first benchmark for real-world visuals with 7M+ creators and counting. Made by @intelligence_ai
Joined June 2025
- Tweets827
- Following16
- Followers25.2K
- Likes1.8K
Pinned Tweet
Design Arena is excited to introduce the company behind our work: Intelligence.
Over the past year, millions of people have helped us evaluate how well AI models perform on real-world creative tasks - where quality is subjective, outcomes are difficult to verify, and human judgment matters.
Design Arena is only the beginning. Let us know what you want to see next!
Today, we're introducing @Intelligence_ai.
In 6 months, as a team of 10, we scaled from $5M to $60M ARR and 5.5M users across 190+ countries.
We raised a $7.9M seed, led by @IndexVentures with participation from @conviction, @A_StarVC, and @combinator to build DesignArena, a universal interface for accessing and evaluating the world's AI capabilities.
Most evaluations try to simulate the real-world. We believe the real-world is the ultimate verifier.
People come to @DesignArena with a request. Models compete to fulfill the request, and users determine what works best for them.
Their live user behavior evaluates the models, improves how work is routed, and helps people access the right intelligence.
We've helped the world's leading frontier labs break the news on their SOTA capabilities.
What's the limit? Join us and find out.
Models can seemingly degrade over time.
The following scenes were created with the same prompt and the same model, @GoogleAI's Gemini 3.8 Flash, by a user on DesignArena.
The first was generated just 3 days after release, while the second came almost a month later.
The difference is striking. The first scene features a finished bungalow with terraced gardens, while the second is barely recognizable.
Digging into the code, we find that the second scene still attempts to build the garden, windows, and patio, but buries much of it beneath its own terraces.
In aggregate, the share of objects never visible on camera jumps from 18% to 57%. For bushes, the share placed underground jumps from 0% to 100%, and even the house walls go from fully above ground to 64% below it.
Both scenes share the same simple user request, yet a few weeks after release, the model struggles to fulfill it. Try making your own generations and judge for yourself at DesignArena.ai!
Introducing Code Art Arena: perhaps the most meaning-preserving form of generative art.
In this arena, models write p5.js programs that draw stroke by stroke, and without ever seeing the canvas they're creating. They must plan and compose the entire artwork through code. Because every piece is code, it's resolution-independent, seedable, and editable.
Taken to its limit, we believe Code Art could be one of the most meaning-preserving ways to generate art. Unlike diffusion models, which iteratively refine an image from noise, Code Art explicitly specifies the sequence of strokes and shapes that construct it - much like a human artist.
Leaderboard coming soon! In the meantime, here are some recent generations:
BREAKING: Kimi K3.1 Pro by @Kimi_Moonshot is projected to rank #1 among open-weights models on DesignArena's Website leaderboard, and is tied for #3 among all models.
Kimi K3.1 Pro has a 15 Elo point lead on their previous model, Kimi K3, on the open weights leaderboard. Among all models it ranks behind only Muse Wonder (max) and Muse Spark 1.3 (xhigh). That ties it with Muse Spark 1.3 Max for third place on the overall Website leaderboard.
Congrats to the @Kimi_Moonshot team on the launch!
Kimi K3.1 Pro by @Kimi_Moonshot is now available on Design Arena!
Kimi K3.1 Pro is built for complex coding, agentic workflows, and multimodal tasks. It can develop full projects in Unreal, Unity, and Godot, tackle parametric 3D modeling in Blender and CAD, and improve financial modeling accuracy with 65% fewer hallucinations.
Congrats to the @Kimi_Moonshot team on the launch!
@OpenAI's GPT-6 Astra ignores your images. That just might be its superpower.
Users upload a reference image and ask for a website. Other frontier models take it literally and "mirror" the input, reusing the same colours and layout and sometimes pasting the image right in.
GPT-6 Astra takes the idea and builds it in its own style. And users still pick it.
This even holds across generations.
From GPT-5.5 to GPT-5.6 Sol to GPT-6 Astra, each GPT release has mirrored the user's image less and projected more of its own house style. Max effort pushes Astra further still.
How can we check that this mirroring is a real effect and not the result of noise?
By scrambling the images, we give every website a random reference image from the same model's set and measure again.
Whereas under real images the 10 newest models score 0.21 to 0.31, after scrambling, they all each score about zero.
So when GPT-6 Astra strays from your image, it does so consistently. Whether that's taste or stubbornness, you can judge for yourself at DesignArena.ai!
DesignArena retweeted
Canva’s first-ever image model has debuted at #4 on @DesignArena
This makes @canva the #2 AI image lab in the world, behind only @OpenAI
An incredible achievement from a very hardworking and talented team pushing the boundaries of what design companies accomplish in frontier AI research
Huge congratulations to @MelanieCanva and the rest of the @canva team!
Canva Craft, the first ever image model by @canva, takes the #4 spot on DesignArena's Image leaderboard. This makes Canva the #2 Image Lab after @OpenAI!
Notably, it wins most often on photorealistic product and object shots, setting up studio-style lighting and detailed material textures on a single hero object. It also ranks #1 on the Portrait and Landscape subcategories.
Congrats to the @canva team on the launch!
DesignArena retweeted
Canva Craft debuted as the world's 4th most advanced image generation model on @DesignArena with an Elo score of 1340.
It delivers exceptional visual quality and state-of-the-art text rendering, outperforming some of the world's leading frontier models.
DesignArena retweeted
Introducing Canva Craft 🚀
Extraordinary image generation is now available to everyone, powered by our first frontier model built specifically for design.
With Craft, Canva is now the world's second most advanced AI lab for image generation on @DesignArena's global leaderboard.
See what it can do 👇
Canva Craft, the first ever image model by @canva, takes the #4 spot on DesignArena's Image leaderboard. This makes Canva the #2 Image Lab after @OpenAI!
Notably, it wins most often on photorealistic product and object shots, setting up studio-style lighting and detailed material textures on a single hero object. It also ranks #1 on the Portrait and Landscape subcategories.
Congrats to the @canva team on the launch!
Our paper, AudioAgentBench, was accepted to the @NeurIPSConf 2026 Evaluations & Datasets Track!
We’ve open-sourced a suite of six static, multi-turn benchmarks designed to stress-test speech-to-speech models on realistic customer-facing tasks. Here’s what we built:
5/ Conversation length alone didn’t explain degradation. On a 30-turn grocery task, tool-use accuracy fell 40.5 points between the first and second halves. A 75-turn conference task fell 11.5 points.
We found that how often a task updates shared state matters more than its turn count.
6/ AudioAgentBench builds on Kwindla Kramer’s original 30-turn evaluation at Daily. Huge thanks to the OpenAI Audio team for helping us expand it and its methodology.
- Methodology: audioarena.ai/methodology
- Code: github.com/Design-Arena/audi…
- Data: huggingface.co/datasets/arca…
