Start your day with melty ai bubbles. Runway Gen4: > @ woman holds the @ bottle up, the bottle is the subject, the @ woman is visible but she is blurred, she is in the @ living_room, it is a product photo shoot (Hand incoherencies fixed with a flux-kontext pass)
MULTIMODAL AI
-
AI Comic Generator Tool Maintains Consistency for Long Narratives
By
–
一键生成 AI 漫画，可以保持一致性，长漫画支持，有灵感的完全可以上。 身为「权利的游戏」粉丝的我，一直对里面人物的造型不满意，有创作欲望的人完全可以用类似的工具出片。 官网 https://
comic-ai.ai/zh -
Higgsfield Soul Launches Pixel-Perfect Inpainting for Creative Editing
By
–
Higgsfield Soul just dropped a game-changing update: pixel-perfect inpainting meets stunning Soul aesthetics! Creators, you can now effortlessly edit clothes, hair, and objects; turning static visuals into dynamic, lifelike creations. Find Higgsfield Soul and other https://
x.com/higgsfield_ai/
/higgsfield_ai/status/1940835284104761454
… -
ChatGPT o3 may only see image after blurring via tool use
By
–
Possible, but a few details suggest otherwise: 1. ChatGPT o3, also with search, doesn’t seem to be able to do it (0/3 attempts) 2. It doesn’t work if you just ask “What is this?” 3. The reasoning trace summary suggests it only sees it after blurring (presumably via tool use)
-
Most LLMs can’t see the Mona Lisa, but o3-pro with squint works
By
–
Most LLMs can’t see this is the Mona Lisa, but it works if you use o3-pro and squint:
-
Veo 3 Video Generation: Two Years of AI Progress
By
–
Two years later, quite a difference.
— Ethan Mollick (@emollick) 4 juillet 2025
This is Veo 3, same prompt “The most American thing ever “ (from the first set of videos generated). https://t.co/IqjSAa0x9H pic.twitter.com/wE6lIoKQlATwo years later, quite a difference. This is Veo 3, same prompt “The most American thing ever “ (from the first set of videos generated).
-
Sketching with o3: Transform rough drawings into labeled diagrams
By
–
Aside: I like that I can sketch this badly on paper, take a photo, then o3 can plot it and add the labels
-
Google Gemini Canvas: Interactive Fireworks Display No-Code AI
By
–
Inspired by the annual Fourth of July fireworks and hot dog eating contest in New York City, we used Canvas in the @GeminiApp to build an interactive fireworks display and a game to celebrate—no coding required. 🧵
— Google AI (@GoogleAI) 4 juillet 2025
Create your own fireworks show 🎆: https://t.co/BLszxH7n96 pic.twitter.com/u83KcmeECXInspired by the annual Fourth of July fireworks and hot dog eating contest in New York City, we used Canvas in the @GeminiApp to build an interactive fireworks display and a game to celebrate—no coding required. Create your own fireworks show : https://
g.co/gemini/share/4
e164b5cd2f9
… -
Grok 4 Benchmarks vs Other Models
By
–
Grok 4 early benchmarks in comparison to other models. Humanity last exam diff is Visualised by @marczierer
-
Abacus.AI ChatLLM: All-in-One Platform for Emails, Projects, Images and Videos
By
–
All-in-One AI Platform for Everyday Tasks! Why settle for just chat when you can have it all? http://
Abacus.AI’s ChatLLM helps you draft emails, manage projects, generate images & videos, and even automate workflows.
No technical expertise needed—just your ideas and our