===== @EHuanglu (originals) ===== 11-08 11:29 219L/ 13410V [video ] ep 02 https://t.co/LGx2B0TYqM QT: AI can make 3D animation for Disney now https://t.co/cM4ctSMhC8 11-08 22:25 131L/ 14436V [video ] what if.. claude let you EDIT the world https://t.co/RkwUgdQJkN 11-10 00:19 420L/ 170910V [video ] fable 5.2 can turn any script into Disney movie now.. https://t.co/byjAKzQtQG QT: Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges. They’re capable collaborators on ambitious projects in many fields, and show improvement 11-10 08:55R 8L/ 4555V [- ] just uploaded a script to your own Claude chat it used Seedance 2.5 on https://filmera.ai via MCP, this way you can literally create a full film 11-10 17:56 11L/ 2408V [- ] Early. That’s the honest word for world models right now, and PAN 2’s own stats say so. The difference is this one is open enough to get less early faster. QT: Introducing PAN 2: an open foundation world model for physical intelligence. A language model predicts the next word. PAN 2 predicts the next world: what happens after each action, grounded in time, 11-10 18:35 35L/ 3899V [video ] holy.. Fish Audio's Live 3 can listen and talk at the same time it says "hmm" while you're mid-sentence and waits for you to finish voice AI is getting crazy https://t.co/Z9sTSefMd4 QT: Introducing @FishAudio Live 3: a full-duplex voice model that listens, says "mm-hmm," and knows when to shut up. Enterprise? Book a demo for your free custom agent: https://s.fish.audio/vxhgs6 Rep 11-10 19:18 14L/ 2710V [- ] if your team uses ai video you should try Sunra at 56% less QT: Today we're launching Sunra Enterprise, with team plans up to 56% off API prices. One workspace for your whole team, with every leading model, a shared credit pool, admin controls and priority supp 11-11 02:50 13L/ 2067V [video ] AURA https://t.co/RkePV6vjmn ===== @c_valenzuelab (originals) ===== 11-09 14:02 55L/ 5028V [video ] Tough Love. Tough Models. https://t.co/PesVp5eaZ8 11-10 16:04 15L/ 1298V [photo ] https://t.co/dDBpzHleRu 11-10 17:22 32L/ 2218V [- ] video is the final interface QT: Imagine video that responds to you. In our latest inside view, @Ryan__Stephen designer and prototyper at Runway Labs, exploring how real-time video could change the way we learn, create and collabora 11-10 23:10 24L/ 1943V [- ] Meet me at the intersection of Runway and Codex QT: With the @runwayml plugin in Codex and ChatGPT, you can see how a character from your prototype moves, try a different opening shot for a trailer, or turn a product photo into a video ad. Anthony Mag ===== @CharaspowerAI (originals) ===== 11-08 12:00 36L/ 2499V [video ] The Agent in Minimax Design is surprisingly good at turning almost any image into a subtle FPV-style shot. 👀⚡️ https://t.co/ijqPU1JOdp 11-08 16:00 14L/ 2088V [video ] I’d wishlist this game immediately. 👀🔥 Tiny armored knight. Cursed castle. Spike traps. Lava. Collapsing towers. Then the boss fills the ENTIRE screen… and you beat it by reflecting its own fireball back into its crown. That’s the kind of 2.5D action I want more of. made in @LumaLabsAI Full promp 11-08 19:00 21L/ 2164V [video ] This prompt works because the camera is intentionally TERRIBLE. 💀📱 Bad zoom. Lost focus. Shaky framing. People screaming off-camera. Then you drop a blockbuster-sized creature into the middle of it. That contrast between amateur footage + impossible scale sells the illusion ridiculously well. Try i 11-08 19:00R 2L/ 465V [- ] Ultra-realistic amateur blockbuster creature footage, 25 seconds, filmed on a smartphone in bright daytime. Raw handheld camera, unstable zoom, autofocus hunting, rolling shutter, compression artifacts, people shouting, sirens, no cinematic stabilization. 0-5s: Normal downtown street. The filmer c 11-09 08:00 9L/ 1405V [video ] GM everyone ☀️ New week. Fresh ideas. Way too many things I want to create already. 😂 AI, VFX, motion design, cinematic experiments… Time to turn the creative engine back on. 🎬🔥 https://t.co/8x3cvAQfP6 11-09 12:00 7L/ 1150V [photo ] Your selfie. But you accidentally spawned inside Dragon Ball Z. 💀🔥 That’s it. That’s the prompt. Upload your photo, keep your identity, switch yourself into full DBZ anime mode… and casually take a selfie with Bulma at Capsule Corp. 😂 Best part? You can remix this with basically any anime universe + 11-09 16:00 19L/ 1786V [video ] Running from zombies is easy. Looking COOL while doing it? That's an art form. 🧟‍♂️😂 When the apocalypse hits, I'm not just trying to survive. I'm making sure every escape looks like a Hollywood action sequence. 🎬🔥 #DreaminaCPP @dreamina_ai https://t.co/Mgq3D4GDKz 11-09 19:00 24L/ 1897V [video ] Sometimes the best VFX shot is just ONE impossible moment executed perfectly. ❄️👀 No fight. No destruction montage. No chaos for the sake of chaos. Just a monk standing still while an entire avalanche has to bend around him. Simple concept. Massive scale. I’m sharing the full prompt below — steal i 11-09 19:00R 1L/ 329V [- ] Ultra-realistic epic spiritual fantasy, one seamless 25-second shot. Slow push-in, low-angle hero framing, snow particles, dramatic mountain daylight, premium energy and ice VFX, blockbuster realism. A blind monk meditates in a snowy mountain village as a colossal avalanche rushes toward the settle 11-09 19:18 8L/ 1709V [photo ] First test with Minimax H3 Image in early access! This time, I'm exploring that iconic 90s movie aesthetic: cinematic lighting, film texture and the unmistakable feel of a classic blockbuster. There's something about that era I just can't get enough of. 🔥 #H3Image @Hailuo_AI 11-10 08:00 15L/ 1435V [video ] GM everyone ☀️ Another day to turn caffeine into pixels, VFX and questionable amounts of rendering. 😂🎬 The creative lab is officially open. https://t.co/PS7z444Zv9 11-10 12:00 30L/ 94600V [photo ] Minimax H3 Image is almost here! 👀🔥 I've had the chance to play with the early access version, and here are a few of my first experiments. Really enjoying exploring what this model can do, especially when it comes to bringing different visual styles to life. 🎬 Curious to see what everyone will cre 11-10 16:00 11L/ 967V [photo ] Pick any movie. Turn it into the LEGO collector set you wish actually existed. 🧱🎬 That’s the prompt. Swap [MOVIE TITLE], choose the most iconic vehicle, creature or location… and let GPT Image 2.5 do the rest. This one is dangerously addictive if you’re a movie fan. 😂 Full reusable prompt below 👇 C 11-10 19:00 28L/ 1542V [video ] This is where Seedance 2.5 gets REALLY interesting. 👀 Not another pretty sci-fi shot. A full 30-second survival-horror trailer with actual pacing, tension, dialogue and a controlled reveal. The goal was to make it feel directed, not generated. And Seedance handled the atmosphere way better than I e 11-10 19:00R 0L/ 287V [- ] Ultra-realistic sci-fi horror trailer, 30 seconds, original industrial survival aesthetic. ARRI Alexa 35, anamorphic lenses, cockpit close-ups, shaky crash energy, helmet lights, slow push-ins, wide planetary reveals, subtle handheld tension, practical ship lighting, dense fog, metallic interiors, o ===== @AI__TSUBAKI (originals) ===== 11-08 02:55 42L/ 1968V [video ] Just Three People Dancing Made with Kling 4.0 on @DomoAI_ #DomoAI #DomoAICPP https://t.co/2h0eP2Bj2J 11-09 11:21 189L/ 15713V [photo ] Seedream 5.0 Pro vs 5.5 What changed for me: ・Skin has real texture now, not that glossy AI finish ・Japanese features finally read as Japanese ・Light and air feel like an actual winter afternoon in Tokyo A real step up for portraits #Seedream5_5 #DreaminaAI https://t.co/em4httxU9G 11-09 11:22R 0L/ 1303V [- ] Prompt: A fashion editorial portrait of a beautiful young Japanese woman in her early twenties, a professional model with a delicate oval face, soft double eyelids, clear luminous skin with a natural healthy glow, and glossy dark brown shoulder-length hair with soft layered bangs moving slightly in 11-09 14:27 16L/ 1483V [photo ] The Ultimate Character Sheet Guide Three panels. One visible face. Every choice has a reason. ① Plain #808080 gray background The exact midpoint between black and white. White bleeds into the footage, black crushes the detail. Using the same gray on every sheet also keeps them consistent when you 11-09 14:27R 5L/ 727V [video ] How I make it: The character itself comes from Midjourney or Higgsfield Soul 2.0 / Soul Cinema. All of them produce realistic subjects. Then I turn it into a sheet with GPT Image 2.6. Quality: High or above, 2K or above recommended. The prompt below is tuned for this character. Save the sheet abo 11-09 18:29 15L/ 1294V [photo ] Testing MiniMax H3 Image in early access. Even at 4K, the cost stays remarkably low. @MiniMax_AI #H3Image https://t.co/Ldb99vPYMU 11-09 19:43 275L/ 12805V [photo ] Poster test with MiniMax H3 Image in early access. Big type and photoreal imagery in one layout — and yes, even at 4K, the cost stays remarkably low. @MiniMax_AI #H3Image https://t.co/RYmROEfsvD 11-09 20:42 9L/ 1023V [photo ] Kimono, Castle, Lanterns, and Blue Sky Made with MiniMax H3 Image @MiniMax_AI #H3Image https://t.co/S1Z1ZVkHHW 11-10 00:28 9L/ 1186V [video ] Ninjas Don't Need Stairs Made with Pollo Recast on @itsPolloAI @polloai_creator #PolloAI https://t.co/mKNEveGBpf 11-10 07:56 5L/ 465V [video ] 7-5-3: Shichi-Go-San Long ago in Japan, so many children died young that a child was said to belong to the gods until the age of 7. Odd numbers were considered lucky, so families gave thanks at 3, 5 and 7. Every November, kids still dress in kimono and visit a shrine. Made with PixVerse V6 Max on QT: PixVerse V6 Max just dropped on WaveSpeed! One idea. Multiple shots. A whole story. - Multi-clip generation for scene transitions - Native audio + custom sound effects - Prompt-guided motion & c ===== @AIWarper (originals) ===== 11-09 18:42 252L/ 12828V [photo ] Testing MiniMax H3 Image in early access: - Accepts up to 16 ref images - 1k, 2k, 4k outputs - 21:9 - 9:16 Aspect Ratio range Currently testing it on cinematics and it's quite good. X compression won't do it justice 6272 width x 2688 height @MiniMax_AI #H3Image https://t.co/RtrkOGkteX 11-09 20:59 8L/ 2413V [- ] How’s Fable 5.2 in your early testing? I am likely going to test it for planning purposes while still having opus 5.5 execute 11-10 16:24 62L/ 3255V [photo ] More testing of MiniMax's H3 Image in early access It's quite addicting to just prompt these ideas for previs / moodboarding purposes. @MiniMax_AI #H3Image https://t.co/5YJArlEbA4 11-10 16:58 62L/ 4620V [- ] Looks like it's out of Early Access! Open weights available too. Excited to train some loras on this one. I suspect it will be SOTA QT: 4K image generation in 10 seconds! ⚡ Meet MiniMax H3-Image. 16 million pixel resolution. Stunning realism. Precise prompt following & editing. Up to 16 reference images. All in 10 seconds. And 11-11 02:29 17L/ 1190V [photo ] Qwen4 inbound https://t.co/bZiuNDZ7rg QT: Qwen4 is coming. Deeper knowledge. Richer world understanding. Stronger coding and agentic capabilities. Built for real-world tasks — from complex coding and knowledge work to front-end builds and p ===== @azed_ai (originals) ===== 11-10 16:54 17L/ 3989V [- ] An EU server doesn't guarantee AI sovereignty. A company can host its AI infrastructure in Europe while leaving critical decisions about access, encryption keys, and data handling in someone else's hands TrueFoundry is challenging that assumption with the EMEA launch of its self-hostable TrueFoun QT: TrueFoundry is expanding across EMEA, with London as our regional hub 🇬🇧🇪🇺 https://t.co/gkcwAFI4yE 11-10 17:58 16L/ 3418V [- ] If my boss finds this, I’m cooked “This’ll take a few days” was elite material QT: Dreamina Marketing Studio is now live!✨ Meet Dreamina Marketing Studio, your all-in-one AI workspace for creating professional marketing content. Bring your creative ideas to life, produce marketing 11-10 19:57 13L/ 3572V [photo ] Got a product to promote? One visual is probably not going to cut it 👀 From Nov 12-18, purchase or upgrade to Ultra on CapCut Web and get 7 days of unlimited image generation on select models Create product visuals, explore different campaign concepts, test out new backgrounds, and generate multip 11-11 00:04 10L/ 869V [photo ] Me: I have a very specific video idea Also me: opens the prompt box and forgets every word😭 If you've been there, Dreamina is worth trying. CapCut Pro members get an extra incentive eligible users can claim all three rewards: up to 98% off Basic, up to 10 free Seedance 2.0 generations, and bonus c 11-11 00:04R 1L/ 141V [- ] Next time you have an idea but can't quite put it into words, find a reference that matches the look you're imagining. So just head to Dreamina Web, sign in with your eligible CapCut Pro account, and reveal your reward. Then use your reference to explore a new creative direction! https://dreamina. 11-11 00:04R 0L/ 137V [- ] If you liked this, follow @azed_ai for more AI tips and prompts. Your likes, reposts, and bookmarks on the main post help me keep creating free content! I appreciate your support! QT: Me: I have a very specific video idea Also me: opens the prompt box and forgets every word😭 If you've been there, Dreamina is worth trying. CapCut Pro members get an extra incentive eligible users c 11-11 05:04 9L/ 465V [photo ] Good morning to everyone already working, everyone still having coffee, and everyone struggling to get out of bed, hope you have a lovely day --sref 3652479600 https://t.co/SGHJgp3LbF ===== @Flovaai (originals) ===== 11-09 12:00 41L/ 1024V [video ] A little clay, a lot of character. 🧡 Your clay-style short film tutorial starts here. https://t.co/JnQIydidu4 11-09 13:00 23L/ 835V [video ] A touch of romance in a vibrant, cinematic 2D cartoon style. https://t.co/ypqOI0Cwye 11-10 13:04 30L/ 898V [video ] AI can now create 3D animations. 🎬 In this tutorial, we’ll show you how, step by step. https://t.co/HqraqnCv3Z ===== @Kling_ai (originals) ===== 11-09 03:11 53L/ 6087V [- ] Through a complete energy drink ad, leedain_leader_in explores how Kling 4.0 handles product details, character consistency, and natural dialogue, shifting more of the creative process from editing to directing. https://x.com/i/article/2118632853118562304 11-10 03:54 33L/ 4252V [- ] Let the journey begin. 🌍🎬 #Kling4 #KlingAI QT: 🎬 OiiOii × Kling 4.0 Creative Contest is HERE! 🌍 Every journey tells a story. You call the shots. This time, the theme is “The Journey”. Explore diverse worlds with OiiOii × Kling 4.0, bring every sce 11-10 04:01R 0L/ 1287V [- ] The full video is here 🎬 https://x.com/Kling_ai/status/2117587110152577217 QT: How much life can you find in black? @EmilHeliot , the Copenhagen-born global fashion brand, just released their new Core Uniform collection campaign film, directed by UNVEIL, creative direction by J ===== @openart_ai (originals) ===== 11-09 20:28 37L/ 2681V [animated_gif] As one of @OpenAI early access partners for GPT Image 2.6, we're excited to see OpenArt officially featured in OpenAI's launch announcement. OpenAI highlighted GIFs created with OpenArt, showcasing how creators can bring their ideas to life with animation from a single prompt. We are also offerin 11-10 16:52 10L/ 1104V [photo ] OpenArt is excited to be a featured company at AFM House 2026, part of the American Film Market. We look forward to connecting with film industry professionals and showcasing how OpenArt can empower your next creative project. Come visit us November 11–14, from 9 AM to 6 PM! https://t.co/7PsvwQXhP 11-10 17:02 27L/ 1729V [video ] Seedream 5.5 by ByteDance is live on OpenArt! Turn one product image into a full set of marketing assets, with fewer credits. • Create realistic product shots and lifestyle scenes • Make precise edits while keeping the rest consistent • Generate transparent backgrounds and layered assets • Create m 11-10 17:02R 0L/ 421V [- ] Try it here: https://openart.ai/suite/create-image/byte-plus-seedream-5-5?utm_source=x&utm_medium=organic_social&utm_campaign=social-x-seedream_5.5-na-acq-web ===== @pika_labs (originals) ===== 11-10 02:41 59L/ 8966V [- ] Huge thanks to @shiyi_c98 for her help in bringing Pika H3 Flex inference to AMD GPUs 💜 We’re excited to collaborate with more researchers in the community. If you’re an AMD developer, we’d love for you to contribute to this branch! Check out her work: https://github.com/Pika-Labs/Pika-H3-Flex/tree 11-10 18:13 54L/ 3998V [video ] Seedream 5.5 is incredible. And the Image Studio app on Pika is an ideal place to take advantage of it. —> Split images into 100+ layers and refine each —> Make precise multi-turn edits by pinning notes, or painting an area —> Export transparent pngs to use in other image and video gens Try it: 11-10 18:25R 6L/ 777V [- ] Thanks to the @byteplusglobal team for yet another astounding model. Try it for FREE in Image Studio if you’re a Creator or Fancy subscriber: https://create.pika.art/apps/image-studio?utm_source=x&utm_medium=caption&utm_campaign=261110-seedream_5_5&utm_content=pika_labs&utm_term=post&utm_id=17024973 ===== @runwayml (originals) ===== 11-09 15:14 79L/ 9344V [video ] Introducing Runway Nights. A new series of intimate events for the creative community. Join us on Wednesday, November 18th in New York. ​Space is limited. Request your spot here. https://luma.com/runway-nights-nyc https://t.co/U7BNG9blMy 11-09 19:27 125L/ 17742V [video ] New in Runway, Sightline. Create a 360 panorama of any location, then frame your shots from any angle, with the environment staying consistent in every one. Export your panorama for 3D to use as an environment in your scenes. Get started in App Studio at the link below. https://t.co/098wikwL2k 11-09 19:27R 3L/ 3068V [- ] Get started: https://sightline--rw.runway.studio/ 11-10 16:58 77L/ 9660V [video ] Imagine video that responds to you. In our latest inside view, @Ryan__Stephen designer and prototyper at Runway Labs, exploring how real-time video could change the way we learn, create and collaborate. https://t.co/fFbIgflWty 11-10 16:58R 10L/ 1904V [- ] Resize the window and the scene rearranges with it. Ryan’s experiments explore video that adapts to its frame, from boxes that reshuffle to a ski map that becomes a closer view of a single slope. QT: We’ve been experimenting with responsive generative video interfaces. Real-time video models are reshaping responsive design. A thread 🧵 https://t.co/CjPGWlH3Of 11-10 16:58R 2L/ 1604V [- ] This drawing prototype uses real-time video to animate your sketches while keeping your style. QT: Real-time video models are unlocking new capabilities. A first experiment: a drawing app that animates your sketches as you draw. Under the hood, a video model generates each frame live, turning ever ===== @higgsfield_ai (originals) ===== 11-09 22:23 747L/ 58325V [video ] Fable 5.2 generates insane Three.js videos. As a leading AI video company, we have to say: the progress is super fast. Every pixel in this Bombay bazaar is code. 6 shots, 21.5 seconds. Here’s how 👇 https://t.co/ZYUQDiQXXh QT: Claude Fable 5.2 is LIVE on Higgsfield. 3 days of FREE & UNLIMITED access to all top video and image models in Higgsfield MCP. Anthropic’s new model excels at 3D modeling, visual reasoning, comp 11-10 02:09 801L/ 139820V [video ] We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access to Seedance 2.5, Seedream 5.5 & every other top model. Start your free trial in Higgsfield MCP now. 11-10 04:07 37L/ 9309V [video ] Claude Fable 5.2 makes code look hand-painted. We gave it one 10-second scene and three styles: >Persian miniature >Gouache >Psychedelic watercolour It drew and animated all three in Canvas2D. No image models or 3D. Full workflow below 👇 https://t.co/4hibW7oZNp QT: We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access 11-10 04:07R 2L/ 1114V [- ] The watercolour version mixes overlapping washes and leaves parts of the paper bare. Watch the colour bleed and the drying edges. The paper grain stays still while the paint moves. Add Higgsfield MCP to Claude https://higgsfield.ai/mcp 11-10 04:07R 2L/ 1105V [- ] All three play over the same 10 seconds, so you can compare each shot before picking a look. The blog has the shared brief, all three style prompts and the render setup. Read the full breakdown https://higgsfield.ai/@higgsfield.studio/blogs/one-scene-three-hand-painted-looks-drawn-entirely-in-code 11-10 04:13 102L/ 13149V [video ] We asked Claude Fable 5.2 to put a console-style boss fight in a browser. Here's the result: > About 3,400 lines of JavaScript on Three.js > Characters designed and turned into 3D through Higgsfield MCP > Two combat stances, perfect dodges and three skills > A temple boss waiting at the end of the QT: We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access 11-10 04:13R 5L/ 1379V [- ] The game and source code are free. Download the 112 MB package and run it locally in Chrome using the setup instructions below. Connect Higgsfield MCP. Start with your hero. https://higgsfield.ai/mcp 11-10 04:13R 3L/ 1276V [- ] Full workflow, starter prompts and setup: https://higgsfield.ai/@higgsfield.studio/blogs/this-looks-like-a-console-game-it-s-a-web-page-claude-fable-5-2-coded 11-10 04:17 222L/ 19673V [video ] Claude produced a launch film for a product it invented. Fable 5.2 and up to 10 helpers built the robot joint in Blender, entirely in code. 25 seconds. First final delivered 4h31m after the brief. Then came the making-of. 👇 https://t.co/ljqaUA3HxS QT: We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access 11-10 04:17R 18L/ 4587V [- ] Claude wrote the brief, generated the music on Higgsfield and brought previews back for approval at three stages. Add Higgsfield MCP to Claude https://higgsfield.ai/mcp 11-10 04:17R 5L/ 1570V [- ] The original request, music prompts and the launch-film-operator skill are all in the blog. Read the full breakdown https://higgsfield.ai/@higgsfield.studio/blogs/one-claude-operator-ten-helpers-how-the-torqk-launch-film-was-made 11-10 14:22R 5L/ 2094V [- ] Start your free 72-hour Higgsfield MCP trial with unlimited access to all top video, image and audio models. $0 now. Cancel anytime. We’ll email you before your trial ends. After 72 hours, it auto-renews to Higgsfield Ultra with 40% off your first month unless you cancel. Video models: • Seedance 11-10 15:59R 2L/ 1378V [- ] Full breakdown with all the prompts: https://higgsfield.ai/@higgsfield.studio/blogs/a-bombay-bazaar-at-sunset-rendered-entirely-in-code 11-10 21:36 114L/ 11321V [video ] Seedream 5.5 is LIVE on Higgsfield. ByteDance's newest image model with less detail loss across edits. > Native RGBA output, ready for compositing > Pixel-level local edits > More natural lighting, materials and skin > Sharper small text Available now on Higgsfield. Coming soon to the US. 11-10 21:36R 2L/ 1218V [- ] Try Seedream 5.5 now on Higgsfield. Coming soon to the US. https://higgsfield.ai/ai/image?model=seedream_5_5 ===== @perplexity_ai (originals) ===== 11-09 19:29 409L/ 35616V [photo ] Claude Fable 5.2 is now available in Perplexity Computer for Pro and Max subscribers, and in Search for Max subscribers. On our WANDR benchmark, it achieved Pareto frontier of performance and efficiency, with a record score of 0.851 at $14.51 per task. https://t.co/P510QumCGW 11-10 17:39 386L/ 29426V [photo ] Gemini 4 Carbon is now available in Perplexity Computer for Pro and Max subscribers, and in Search for Max subscribers. It scores 0.857 on our WANDR benchmark at $5.86 per task, matching Claude Fable 5.2 at 40% of the cost. https://t.co/4gGZc3fSJB ===== @ElsaSofia__AI (originals) ===== 11-08 02:11 342L/ 44679V [video ] This is what the next generation of AI video looks like. Generated with Seedance 2.5 Prompt below:⤵️ https://t.co/qiHlNSuYN0 11-08 02:11R 13L/ 1542V [photo ] THE BLONDE WOMAN WITH THE BLACK UMBRELLA Same-to-Same Cinematic Video Prompt Duration: 30 seconds Aspect Ratio: 16:9 Style: Photorealistic live-action, surreal street comedy, cinematic urban storytelling. CHARACTER CONSISTENCY Main Woman: A stylish young adult blonde woman with long, naturally 11-09 02:01 154L/ 7173V [video ] The level of realism in AI video creation is truly remarkable. Created with Seedance 2.5 Prompt below:⤵️ https://t.co/B9vme6BaGm 11-09 02:01R 4L/ 768V [photo ] THE GIRL SUMMONS A METEOR — CINEMATIC FANTASY COMEDY Duration: 17 seconds Aspect Ratio: 16:9 Style: Photorealistic live-action Korean fantasy comedy, cinematic VFX, realistic human movement. CHARACTER CONSISTENCY Female Character: A beautiful young Korean schoolgirl with long, loose, wavy blonde 11-09 13:26 82L/ 14136V [- ] this guy literally showed everyone how to make UNLIMITED videos for FREE... and somehow people are STILL sleeping on it. QT: https://x.com/i/article/2119061058480644096 11-09 14:10 103L/ 9019V [video ] Creating content is one thing. Sharing it across multiple platforms is another. YORK AI helps adapt your existing content for X, Instagram, Facebook, and other supported platforms. You can review each version, approve your favorites, and organize your publishing schedule in one place. @_YorkAI QT: Introducing York AI, your AI Social Media Manager. Your entire social media operation, on autopilot. From spotting trends and creating content to scheduling posts, managing engagement, and analyzin 11-10 16:15 49L/ 710V [- ] A new SOTA image model has arrived in the open-weight AI ecosystem! MiniMax H3 Image delivers impressive performance, and its availability with open weights is a big step for the AI community. #H3Image QT: 4K image generation in 10 seconds! ⚡ Meet MiniMax H3-Image. 16 million pixel resolution. Stunning realism. Precise prompt following & editing. Up to 16 reference images. All in 10 seconds. And 11-10 17:49 66L/ 767V [- ] A seasonal fashion campaign needs a direction before it needs another prompt. @hypitai makes that starting point refreshingly clear. Choose a reference ad. Add the product you want to feature. Ask your agent to rebuild the creative around your collection and selling points. Hypit works through th QT: We just killed Higgsfield Ad Multiplier. Meet the world’s #1 viral ad cloner. Drop in your competitor’s best-performing ad and a photo of your product. Hypit rebuilds it around your product, then gen ===== GLOBAL MODEL / PRICE GREPS (in-window originals) ===== corpus 360 -- \bveo\b: 0 -- -- \bsora\b: 0 -- -- \bseedance\b: 11 -- @higgsfield_ai 11-10 02:09 801L | We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access to Seedance 2 @ElsaSofia__AI 11-08 02:11 342L | This is what the next generation of AI video looks like. Generated with Seedance 2.5 Prompt below:⤵️ https://t.co/qiHlNSuYN0 @ElsaSofia__AI 11-09 02:01 154L | The level of realism in AI video creation is truly remarkable. Created with Seedance 2.5 Prompt below:⤵️ https://t.co/B9vme6BaGm @CharaspowerAI 11-10 19:00 28L | This is where Seedance 2.5 gets REALLY interesting. 👀 Not another pretty sci-fi shot. A full 30-second survival-horror trailer with actual pacing, tension, dialogue and a controlled reveal. The goal was to mak @AiWithTariq 11-10 03:52 25L | This AI video is absolutely a masterpiece 🔥 Seedance 2.5 https://t.co/CT3D6Wtzwq @higgsfield 11-10 14:22 24L | Start your free 72-hour Higgsfield MCP trial with unlimited access to all top video, image and audio models. $0 now. Cancel anytime. We’ll email you before your trial ends. After 72 hours, it auto-renews to Hi -- \bkling\b: 4 -- @Kling_ai 11-09 03:11 53L | Through a complete energy drink ad, leedain_leader_in explores how Kling 4.0 handles product details, character consistency, and natural dialogue, shifting more of the creative process from editing to directing @AI__TSUBAKI 11-08 02:55 42L | Just Three People Dancing Made with Kling 4.0 on @DomoAI_ #DomoAI #DomoAICPP https://t.co/2h0eP2Bj2J @AITOPIAai 11-10 13:06 23L | We asked our Gen Z designer to make a Kling 4.0 ad. This is what they sent back. Do we give them a raise? https://t.co/0POj5wVCOD @AITOPIAai 11-10 13:07 2L | At least they got the announcement right. Kling 4.0 is now on AITOPIA. Try it: https://aitopia.ai/kling/v4 -- \bhailuo\b: 0 -- -- per (second|sec): 0 -- -- /sec\b: 0 -- -- \bpricing\b: 2 -- @OpenAI 11-10 18:06 478L | Use Business Hub to shape what customers see about your business in ChatGPT. Once verified, you can create a Business Agent, update your business details, and add product feeds with current pricing and availab @AITECHio 11-09 16:00 127L | $0.80 an hour. Know your GPU compute costs. AITECH’s published starting rate for an NVIDIA L40S is $0.80 per GPU card per hour.* Configurations and providers vary, so compare the GPU model, available capacity -- \bcost\b: 8 -- @hubermanlab 11-09 15:00 758L | The new Huberman Lab episode is out: Overcome Addictions and Obsessions of All Kinds | Dr. Anna Lembke 0:00 Anna Lembke 2:14 Social Media, Digital Media Addiction, 4 Cs; Pollution Analogy 6:43 Dopamine, Reinfo @AITECHio 11-10 08:00 413L | Five tools, one workflow, constant context-switching. The tools aren’t the cost. The switching is. Every switch costs a few seconds of re-orientation. Where was I? What did I just see? What was I about to do? @perplexity_ai 11-10 17:39 386L | Gemini 4 Carbon is now available in Perplexity Computer for Pro and Max subscribers, and in Search for Max subscribers. It scores 0.857 on our WANDR benchmark at $5.86 per task, matching Claude Fable 5.2 at 40 @AIsaOneHQ 11-09 05:08 371L | Introducing AIsa Social. Your agent can now turn social insights into action: > Finds viral videos on TikTok, Instagram and X > Picks a topic and creates content for each platform > Schedules and publishes to @AITECHio 11-10 16:00 338L | $1,728. That’s an H100, running non-stop, for a month. That’s the estimated cost of always-on compute at our published starting rate, assuming 24/7 usage for 30 days. Most workloads don’t run a GPU around the @AI__TSUBAKI 11-09 19:43 275L | Poster test with MiniMax H3 Image in early access. Big type and photoreal imagery in one layout — and yes, even at 4K, the cost stays remarkably low. @MiniMax_AI #H3Image https://t.co/RYmROEfsvD -- \bcredits?\b: 4 -- @DiscussingFilm 11-11 02:10 23584L | A new teaser scene for ‘AVENGERS: DOOMSDAY’ is attached to the post-credits of ‘VISIONQUEST’. https://t.co/Xnapl0FQR6 @openart_ai 11-10 17:02 27L | Seedream 5.5 by ByteDance is live on OpenArt! Turn one product image into a full set of marketing assets, with fewer credits. • Create realistic product shots and lifestyle scenes • Make precise edits while ke @azed_ai 11-11 00:04 10L | Me: I have a very specific video idea Also me: opens the prompt box and forgets every word😭 If you've been there, Dreamina is worth trying. CapCut Pro members get an extra incentive eligible users can claim a @AIFrontliner 11-10 08:34 0L | Stop burning credits and follow this instead. -- \bopen[\s-]?weights?\b: 2 -- @AIWarper 11-10 16:58 62L | Looks like it's out of Early Access! Open weights available too. Excited to train some loras on this one. I suspect it will be SOTA @ElsaSofia__AI 11-10 16:15 49L | A new SOTA image model has arrived in the open-weight AI ecosystem! MiniMax H3 Image delivers impressive performance, and its availability with open weights is a big step for the AI community. #H3Image -- \bminimax h3\b: 11 -- @AI__TSUBAKI 11-09 19:43 275L | Poster test with MiniMax H3 Image in early access. Big type and photoreal imagery in one layout — and yes, even at 4K, the cost stays remarkably low. @MiniMax_AI #H3Image https://t.co/RYmROEfsvD @AIWarper 11-09 18:42 252L | Testing MiniMax H3 Image in early access: - Accepts up to 16 ref images - 1k, 2k, 4k outputs - 21:9 - 9:16 Aspect Ratio range Currently testing it on cinematics and it's quite good. X compression won't do it @ElsaSofia__AI 11-10 16:15 49L | A new SOTA image model has arrived in the open-weight AI ecosystem! MiniMax H3 Image delivers impressive performance, and its availability with open weights is a big step for the AI community. #H3Image @CharaspowerAI 11-10 12:00 30L | Minimax H3 Image is almost here! 👀🔥 I've had the chance to play with the early access version, and here are a few of my first experiments. Really enjoying exploring what this model can do, especially when it c @WuxiaRocks 11-10 10:20 24L | MiniMax H3 Image is great for video game assets. It can produce any style and aesthetic. There's much more to explore here. @Hailuo_AI #H3Image https://t.co/8KrSfkmrBU @WuxiaRocks 11-10 07:00 19L | I've been given early access to the new Minimax H3 Image Model. After a few quick tests, let me tell you, it's a rich model. It can do so many styles. Also, 16 image references! 4K res. Here are my quick test r -- \bseedream 5\.5\b: 9 -- @higgsfield_ai 11-10 02:09 801L | We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access to Seedance 2 @higgsfield_ai 11-10 21:36 114L | Seedream 5.5 is LIVE on Higgsfield. ByteDance's newest image model with less detail loss across edits. > Native RGBA output, ready for compositing > Pixel-level local edits > More natural lighting, materials @pika_labs 11-10 18:13 54L | Seedream 5.5 is incredible. And the Image Studio app on Pika is an ideal place to take advantage of it. —> Split images into 100+ layers and refine each —> Make precise multi-turn edits by pinning notes, or p @openart_ai 11-10 17:02 27L | Seedream 5.5 by ByteDance is live on OpenArt! Turn one product image into a full set of marketing assets, with fewer credits. • Create realistic product shots and lifestyle scenes • Make precise edits while ke @higgsfield 11-10 14:22 24L | Start your free 72-hour Higgsfield MCP trial with unlimited access to all top video, image and audio models. $0 now. Cancel anytime. We’ll email you before your trial ends. After 72 hours, it auto-renews to Hi @higgsfield_ai 11-10 14:22 5L | Start your free 72-hour Higgsfield MCP trial with unlimited access to all top video, image and audio models. $0 now. Cancel anytime. We’ll email you before your trial ends. After 72 hours, it auto-renews to Hi -- \bgemini 4\b: 2 -- @GoogleAI 11-10 16:12 5789L | Our strongest frontier model, Gemini 4 Carbon, is now available. Built for deeply complex, long-horizon workflows that push a frontier model’s intelligence, Carbon delivers frontier performance in agentic work @perplexity_ai 11-10 17:39 386L | Gemini 4 Carbon is now available in Perplexity Computer for Pro and Max subscribers, and in Search for Max subscribers. It scores 0.857 on our WANDR benchmark at $5.86 per task, matching Claude Fable 5.2 at 40 -- \bfable 5\.2\b: 15 -- @AnthropicAI 11-09 19:00 14771L | Claude Fable 5.2 and Claude Mythos 5.2 are available now: @higgsfield 11-09 20:06 2832L | Claude Fable 5.2 is LIVE on Higgsfield. 3 days of FREE & UNLIMITED access to all top video and image models in Higgsfield MCP. Anthropic’s new model excels at 3D modeling, visual reasoning, computer use and c @higgsfield_ai 11-10 02:09 801L | We asked Claude Fable 5.2 to finish Leonardo da Vinci's flying machine. It built a 3D model, stress-tested it under real-world physics and redesigned it to fly. 3 days of FREE & UNLIMITED access to Seedance 2 @higgsfield_ai 11-09 22:23 747L | Fable 5.2 generates insane Three.js videos. As a leading AI video company, we have to say: the progress is super fast. Every pixel in this Bombay bazaar is code. 6 shots, 21.5 seconds. Here’s how 👇 https://t. @EHuanglu 11-10 00:19 420L | fable 5.2 can turn any script into Disney movie now.. https://t.co/byjAKzQtQG @perplexity_ai 11-09 19:29 409L | Claude Fable 5.2 is now available in Perplexity Computer for Pro and Max subscribers, and in Search for Max subscribers. On our WANDR benchmark, it achieved Pareto frontier of performance and efficiency, with -- \bnano banana\b: 2 -- @higgsfield 11-10 14:22 24L | Start your free 72-hour Higgsfield MCP trial with unlimited access to all top video, image and audio models. $0 now. Cancel anytime. We’ll email you before your trial ends. After 72 hours, it auto-renews to Hi @higgsfield_ai 11-10 14:22 5L | Start your free 72-hour Higgsfield MCP trial with unlimited access to all top video, image and audio models. $0 now. Cancel anytime. We’ll email you before your trial ends. After 72 hours, it auto-renews to Hi ===== HEADLINE DELTAS (clean) ===== @venturetwins 2119846806066733206 likes 23972->27610 (+3638) views 3016221->8164575 (+5148354) rate 0.338% @higgsfield 2119884452054937656 likes 1663->2832 (+1169) views 260942->700488 (+439546) rate 0.404% @AnthropicAI 2119867707546071125 likes 13099->14771 (+1672) views 654106->865352 (+211246) rate 1.707% @EHuanglu 2119947980400976267 likes 128->420 (+292) views 15391->170910 (+155519) rate 0.246% @higgsfield_ai 2119975721909301684 likes 304->801 (+497) views 28037->139820 (+111783) rate 0.573% @venturetwins 2119579582592630873 likes 5171->5197 (+26) views 233135->239119 (+5984) rate 2.173% @venturetwins 2119993792875180244 likes 376->3784 (+3408) views 13866->130263 (+116397) rate 2.905% unique @handles in md: 75 | scanned 61 in md not scanned: ['4aibsc', 'ai_filmmaker', 'ai_video_dumps', 'airportaction', 'chandan_perla', 'corridordigital', 'jeffsynthesized', 'jiamingsuen', 'magiermogul', 'moonvalley', 'mystageai', 'paultrillo', 'soaima_ai', 'visiblemaker']