🚨 Gemini 3.0 Pro - ecpt checkpoint Holy shit Guys , i want everyone to see this retweet as much as you can to get this to mainstream , i dont ask for this normally All apps work , apple animation , minimize , tools , browser , and everything literally this is the best we can

🚨 Open AI Gpt-5.2 Now it's time for a new test and comparison, comment which models everyone wants to see to be compared of

https://t.co/qDpgbnNV7Y

https://t.co/bJ4xnQPX95

How To Test AI model coding and multimodality at the same time? Answer is animated svg Best svg means Best model and vice versa For the @yupp_ai competition i want you all to focus on what you see in your daily life 1. t can be rick and morty , nintendo , ps5 , cats and

Prompt : A photo of an everyday scene at a busy cafe serving breakfast. In the foreground is an anime man with blue hair, one of the people is a pencil sketch, another is a claymation person

🚨 GPT-image-2 on Designarena Holy Shit it cooked hard 😱 Chestnut can combine different styles in a single image. In this photo we've included an anime character, a sketched character and a claymation one, all with just a prompt (provided in comment)

Still that yellow tint cmon bro solve this competition is way ahead same prompt different results Hazelnut {left} SeeDream-4 {right}

🤌Image made by chestnut prompt : A close-up image of the character Kakashi Hatake facing forward with an imposing and cold expression. The background should be dark and shadowy, creating a mysterious atmosphere. The character should occupy most of the image, with his iconic

🚨 Breaking News Both these mystery models maybe an open ai image model here is the proof > hazelnut is bigger model > chestnut is small and weaker prompt is in binary to test reasoning and they aced it

prompt : 01110111 01110010 01101001 01110100 01100101 00100000 01101111 01101110 00100000 01100001 00100000 01110111 01101000 01101001 01110100 01100101 01100010 01101111 01100001 01110010 01100100 00100000 01110111 01101000 01101001 01100011 01101000 00100000 01100011 01101111

code pen link

😱 Gemini DeepThink is insane Made a saas website fully voice controlled and gesture control and it took two prompts literally This is the first time there is just limit to creativity and marketing , you can just make product the way you want pure independece ,

🚨 Breaking News Two Mystery Image Model Spotted in designarena > Hazelnut > Chestnut

https://t.co/0JGTZ9qLyz

🚨 CAT SVG by GLM 4.6 @Zai_org you guys cooked hard , i have to test more to see how it performs , this was my first test and its insane 🫡 Btw guys , i have seen lots of people making insane svg in our community , I want to announce that I am Judge of @yupp_ai competition

Gemini DeepThink is Insane 🫡 You can generate a website with planets that can be control by hand gesture, it detects hand gestures using your camera and interact this is crazy it took one shot . This was created by one simple prompt thats it , btw i was shy sharing video there

Lithiumflow is better than Deepthink and they made deepthink lazy like it doesnot output longer codes anymore Same Prompt Different Results , cat svg is good i will share it next if everyone want to see

Holy Shit Gemini DeepThink is Insane 🔥 One shotted this website , like bro it added so much stuff in one go and i use the same prompt that i tried with other ai , this does not look like ai slop @OfficialLoganK @GeminiApp congrats guy this is literally out of this world 🫡

👀Introducing a brand new @yupp_ai SVG leaderboard ranking frontier models on the generation of coherent and visually appealing SVGs! Gemini 3 Pro by @GoogleDeepMind takes the crown as the most powerful model! 👏 We’re also releasing a public SVG dataset. Details in🧵