| 1 | Oct 1, '24 | Smart glasses make it possible to unknowingly record people's faces and gather publicly available data like name, home address, phone number, relatives and more all automatically οΏ½ | 540Γ | 8.1M | 5.2k | 852 | 5.1k |
| 2 | Mar 2, '25 | oh btw ai can now generate high-quality classical sheet music and it sounds absolutely insanely good | 190Γ | 2.8M | 4.6k | 466 | 3.7k |
| 3 | Aug 9, '24 | Sorry guys, this time it's so over for real π | 130Γ | 1.9M | 3.4k | 293 | 2k |
| 4 | Mar 26, '24 | Less than 24 hours ago, OpenAI released the first glimpse of what Sora looks like in the hands of artists. This is what unchained creativity looks like! 7 beautiful examples: | 126Γ | 1.9M | 3k | 355 | 2.3k |
| 5 | Sep 16, '23 | GN dreamers β¨ May the spiral bring you peace π΄ | 104Γ | 1.6M | 20k | 2.1k | 1.4k |
| 6 | Mar 17, '25 | most people don't realize the impact of truly multimodal ai models that can generate more than just text these are the real startup killers don't believe me? all of the following t | 81.1Γ | 1.2M | 5.9k | 386 | 6.6k |
| 7 | Dec 2, '23 | This doesn't just work with realistic humans, also with cartoon/anime characters. | 77.2Γ | 1.2M | 1.8k | 178 | 475 |
| 8 | Feb 27, '25 | all that's missing now is some ai rendering that turns the retro blocky style into hyperrealism and everybody and their mom will be able to make AAA games | 50.1Γ | 751k | 1.9k | 83 | 342 |
| 9 | Oct 11, '24 | @NathanLands Last time I visited NYC in 2019 I thought I went to a third world country | 48.3Γ | 725k | 3.8k | 74 | 50 |
| 10 | Dec 1, '23 | High quality AI generated talking heads are coming! GAIA can generate talking avatars from a single portrait image and speech clip. It even supports text prompts like `sad`, `open | 44Γ | 660k | 1.5k | 319 | 1k |
| 11 | Jul 12, '24 | AI lets you hug your dead loved ones again π | 41.1Γ | 616k | 2.8k | 391 | 1.9k |
| 12 | Nov 4, '23 | Exactly 44 hours ago, 12 hours before Midjourney released the Style Tuner and 100 hours before OpenAI is gonna blow our collective minds, Runway released some major improvements to | 37.1Γ | 557k | 1.4k | 190 | 1.4k |
| 13 | Nov 2, '24 | Weekly AI art pump thread! Show me your work. No words. Just vibes β¨ | 35.5Γ | 533k | 12k | 534 | 2.7k |
| 14 | Dec 24, '24 | google deepmind is working on braindance tech | 35.3Γ | 530k | 6.3k | 441 | 1.8k |
| 15 | Dec 2, '23 | High quality AI generated human videos are coming! Animate Anyone can generate videos of anyone with a single image and a bit of pose guidance π€― | β‘ 35.3Γ | 51.2M | 9.8k | 1.9k | 11k |
| 16 | Dec 20, '24 | it's so over you can't even comprehend it this is new "research" from bytedance 100% artificial podcasters will soon exist 7 crazy examples: | 34.4Γ | 516k | 2.6k | 321 | 2.7k |
| 17 | Mar 10, '24 | @Robert_E_Kelly I always lose it when the second one comes in and absolute madness breaks loose π Such a gem of a video. | 33.9Γ | 508k | 5.3k | 39 | 34 |
| 18 | Sep 15, '23 | Felt inspired by the work of @MrUgleh and tried to recreate the effect. Pretty cool. | 33.2Γ | 498k | 6k | 554 | 1.4k |
| 19 | Nov 21, '23 | The Stable Video Diffusion model just dropped π₯ The new model supports: β Text-to-Video β Image-to-Video β 14 or 25 frames at 576 x 1024 β Multi-View Generation β Frame Interpolat | 30.5Γ | 458k | 1.5k | 262 | 890 |
| 20 | Sep 22, '23 | OBEY ποΈ | 28.6Γ | 428k | 4.5k | 771 | 730 |
| 21 | Sep 17, '23 | GN dreamers β¨ I've heard you all like spirals now ππ | 26.7Γ | 401k | 7.2k | 945 | 349 |
| 22 | Sep 23, '23 | Tired of watermarks in videos? ProPainter is a new AI video inpainting method that is able to remove objects, complete masked videos, remove watermarks and even supports video outp | 25Γ | 374k | 2.4k | 370 | 1.8k |
| 23 | Sep 19, '23 | A spiral can take many different forms κ© | 24.7Γ | 370k | 4.2k | 353 | 250 |
| 24 | Sep 18, '23 | Things are spiraling out of control π Since my first spiral post last Friday there have been 2000 of you clicking that Follow button π€― Love our community! Please introduce yourse | 24.5Γ | 368k | 4.5k | 360 | 368 |
| 25 | Apr 22, '24 | Adobe presents VideoGigaGAN! A new video upscaling model that can upsample a video up to 8x with rich details. 10 wild examples β¬οΈ | 22.2Γ | 334k | 1.4k | 220 | 1k |
| 26 | Aug 6, '24 | Made this Claude Artifact today that can generate color palette grids which can be fed to Midjourney as a style reference image β¨ | 22.1Γ | 331k | 1.8k | 170 | 2.2k |
| 27 | Aug 30, '24 | is this what ilya saw? | 21.3Γ | 319k | 1.3k | 79 | 686 |
| 28 | Mar 29, '25 | > be openai > release cool product update > people resubscribe > everybodyβs having fun > wait 24 hours > tweet about overwhelming demand > introduce restrictions > people complain | 21.2Γ | 318k | 3.7k | 233 | 430 |
| 29 | Oct 22, '23 | 3D Gaussian Splats are cool, but they're static (PART 2). 4K4D can turn 4k videos into dynamic scenes and render them in real-time at 80fps on a RTX 4090 GPU. For 1080p videos, the | 20.6Γ | 309k | 1.8k | 268 | 800 |
| 30 | Aug 5, '25 | all the disbelievers last week who said ai will never be able to achieve the omw style and that it's just a 5 second slop video and now deepmind dropped a world model on a regular | 19.8Γ | 297k | 3.2k | 146 | 843 |
| 31 | Jun 22, '25 | i'm still amazed that this is even possible | 19.3Γ | 290k | 3.1k | 285 | 1.4k |
| 32 | Sep 22, '23 | Spiral tutorial time π In this thread I'm going to guide you through the steps to create your own spiral / optical illusion art with Stable Diffusion and ControlNet. Let's jump in | 18.4Γ | 276k | 1.8k | 188 | 2k |
| 33 | Aug 15, '24 | Man, Florida is crazy | 18.4Γ | 276k | 2.1k | 170 | 519 |
| 34 | Nov 10, '23 | Adobe is entering the image-to-3D game! Their new method, LRM, can create high-fidelity 3D meshes from a single image in just 5 seconds π₯ It's trained on 1 million objects and is | 18.1Γ | 271k | 1k | 190 | 408 |
| 35 | Oct 10, '25 | adobe cooked with this | 17.6Γ | 264k | 2.3k | 105 | 1.1k |
| 36 | Dec 2, '23 | RIP TikTokers π | β‘ 17.4Γ | 43.1M | 8.8k | 1.3k | 4.1k |
| 37 | Dec 6, '24 | Thanks to AI, so many amazing new possibilities are emerging. Oh, what I would give to be a kid again. Our imagination will truly be the limit. | 17.3Γ | 259k | 2.7k | 310 | 1.5k |
| 38 | May 24, '25 | bytedance just released BAGEL! an open-source multimodal alternative to gpt-4o and gemini flash it supports: > image reasoning > image editing > video generation > styl | 17.2Γ | 258k | 1.5k | 199 | 1.6k |
| 39 | Feb 14, '25 | oh shit, here we go again! Animate Anyone 2 can replace people in videos with anyone from a single reference image π₯ | 16.9Γ | 253k | 1.8k | 261 | 1.8k |
| 40 | Aug 21, '24 | KEEP can enhance video face super-resolution by maintaining consistency across frames! It uses Kalman filtering to improve facial details, working well on both synthetic and real-w | 16.9Γ | 253k | 1.3k | 172 | 948 |
| 41 | Jul 27, '25 | "ok grok, restore the roman empire to its former glory. make no mistakes." | 16.3Γ | 245k | 5.1k | 299 | 367 |
| 42 | Mar 8, '25 | look at me, i'm the 3d artist now! ai can now create 3d meshes from sketches a child could draw | 16.3Γ | 244k | 2k | 205 | 1.7k |
| 43 | Feb 12, '24 | TikTok presents Boximator! This method can generate rich and controllable motions for image-to-video generations by drawing box constraints and motion paths onto an image and combi | 15Γ | 226k | 1.4k | 302 | 1k |
| 44 | Nov 26, '23 | 3D Gaussian Splats are cool, but they're static (Part 7). PhysGaussian is a simulation-rendering pipeline that can simulate the physics of Gaussian Splats while simultaneously rend | 14.3Γ | 214k | 1.3k | 207 | 496 |
| 45 | Jan 6, '24 | Soon anybody can be animated everywhere! En3D can generating 3D humans either from text or images and animate them! | 13.1Γ | 197k | 1.6k | 326 | 1.2k |
| 46 | Sep 23, '24 | SPARK can create high-quality 3D face avatars from regular videos and track expressions and poses in real time. It improves the accuracy of 3D face reconstructions for tasks like a | 12.9Γ | 194k | 1.7k | 252 | 1.5k |
| 47 | Sep 25, '23 | Portfolio idea π‘ Use the controlism method to link your social media profiles instead of boring plain text. | 12.7Γ | 191k | 2k | 209 | 604 |
| 48 | Oct 30, '23 | Generative 3D is getting better by the week! DreamCraft3D creates high-quality 3D objects from a single text prompt. It first uses a 2D reference image to guide the sculpting then | 12.7Γ | 190k | 1.1k | 154 | 642 |
| 49 | Mar 28, '25 | re-watching your favourite movies gonna be a bit different in a few years | 11.8Γ | 178k | 2.3k | 141 | 614 |
| 50 | Dec 24, '24 | new virtual try-on method, this one comes with code π | 11.7Γ | 175k | 1.1k | 79 | 1.5k |
| 51 | Oct 6, '24 | Nvidia presents EdgeRunner! The method can generate high-quality 3D meshes with up to 4,000 faces at a spatial resolution of 512 from images and point-clouds. | 11.6Γ | 173k | 1.2k | 203 | 746 |
| 52 | Dec 22, '24 | wake up babe, a new upscale method just dropped | 11.4Γ | 172k | 1.2k | 89 | 1.1k |
| 53 | Aug 8, '25 | genie 3 can even let you walk through classic paintings π | 11Γ | 164k | 2.1k | 180 | 707 |
| 54 | Nov 2, '24 | MoGe can turn images and videos into 3D point maps! Links β¬οΈ | 10.8Γ | 162k | 1.8k | 245 | 1.2k |
| 55 | Apr 3, '24 | Meta presents Garment3DGen! A method that can stylize the geometry and textures from 2D image and 3D mesh garments, which can be fitted on top of parametric bodies and be simulated | 10.5Γ | 158k | 1.2k | 216 | 853 |
| 56 | Apr 21, '24 | SAM + Optical Flow = FlowSAM FlowSAM can discover and segment moving objects in a video and outperforms all previous approaches by a considerable margin in both single and multi-ob | 10.5Γ | 158k | 1.2k | 214 | 814 |
| 57 | Feb 22, '25 | not long now before we can simply drag and drop an image into an editor to create a 3D scene! the scenes below are generated from the image on the left π | 10.4Γ | 156k | 1.4k | 153 | 990 |
| 58 | Dec 20, '24 | adobe is cooking a new method to generate sound from vocal imitations π | 10.3Γ | 154k | 1.8k | 224 | 971 |
| 59 | Nov 29, '23 | Say goodbye to green screens! DiffusionMat is an AI framework that uses a diffusion model to create refined alpha mattes by correcting the output at every denoising step to look mo | 9.8Γ | 147k | 1.1k | 198 | 772 |
| 60 | Jan 9, '25 | tencent released a new video face restoration method π₯ | 9.5Γ | 143k | 1.3k | 176 | 778 |
| 61 | Oct 1, '24 | This go you paranoid? Probably time to remove your data from search engines. | 9.5Γ | 142k | 1.4k | 211 | 3.4k |
| 62 | Mar 30, '25 | show me your art bubis. no words. only art π | 9.4Γ | 141k | 1.3k | 45 | 400 |
| 63 | Mar 26, '24 | 1) @shykids | 9.4Γ | 140k | 1k | 57 | 145 |
| 64 | Dec 20, '24 | btw dreamers pika lets you generate unlimited videos for free until the end of the week, even us EU plebs great opportunity to try out the new pika 2.0 scene ingredients tool 5 hig | 9.1Γ | 137k | 1.1k | 86 | 1.2k |
| 65 | Aug 2, '23 | Voice cloning is getting scary. HierVST is zero-shot voice transfer system without any text transcripts. That means it is able to transfer the voice style of a target speaker to a | 9.0Γ | 135k | 1k | 241 | 924 |
| 66 | Sep 18, '23 | GN dreamers β¨ The spiral is everywhere π | 9.0Γ | 135k | 3.2k | 328 | 145 |
| 67 | Jul 7, '24 | AI music videos are so hot right now π₯ | 8.6Γ | 129k | 1.1k | 139 | 592 |
| 68 | Jan 20, '25 | show me your art dreamheads. no words. only art π | 8.0Γ | 121k | 2.5k | 137 | 390 |
| 69 | Mar 26, '25 | during my short time on earth, i've never seen anything like this before | 8.0Γ | 119k | 1.1k | 67 | 267 |
| 70 | Nov 23, '24 | SAMURAI gives SAM 2 motion-aware memory! And the results look mind-blowing. This is zero-shot π€― Links β¬οΈd | 7.8Γ | 117k | 1.1k | 134 | 752 |
| 71 | Nov 11, '24 | Saw this "Text Behind Image" tool today and wanted to see if I can recreate it. And I did. Runs 100% in the browser! First version is live now. Link below π | 7.7Γ | 116k | 1.2k | 122 | 1.4k |
| 72 | Jun 27, '25 | Hunyuan published GameCraft! although we could see for a while now where we are heading, it still blows my mind to see world-simulation models actually take shape i don't think peo | 7.5Γ | 112k | 1k | 107 | 558 |
| 73 | Jun 13, '25 | MiniMax-Remover can remove objects from videos efficiently with just 6 sampling steps: | 7.5Γ | 112k | 1.5k | 181 | 1.2k |
| 74 | Feb 23, '25 | china just dropped Step-Video-T2V, a new high-quality video model π₯ > 30B parameters > text-to-video > 204 frames >quality & turbo mode | 7.3Γ | 109k | 1.1k | 103 | 732 |
| 75 | Jun 18, '24 | 3D Gaussian Splats are cool, but they're static (Part 33). GGHead can generate and render 3D heads at 1K resolution in real-time. | 7.0Γ | 105k | 973 | 147 | 528 |
| 76 | Feb 22, '25 | adobe is cooking a new method that can automatically rig 3D assets by generating joints, skeletons, and skinning weights π | 6.6Γ | 99k | 1.1k | 98 | 468 |
| 77 | Sep 20, '23 | Had some fun during the past few days creating a refined version of my escheresque spiral piece that went viral over the weekend. Added lots of details that are only visible up clo | 6.0Γ | 89k | 1.8k | 163 | 147 |
| 78 | Mar 28, '25 | rip reproductivity π | 5.8Γ | 87k | 1k | 45 | 410 |
| 79 | Apr 11, '25 | James Cameron on AI datasets and copyright: "I think people are looking at it all wrong." "Every human being is a model. You create a model as you go through life." "My output shou | 5.6Γ | 84k | 1k | 154 | 379 |
| 80 | Sep 27, '23 | GN dreamers β¨ May your deepest dreams come true π΄ | 5.3Γ | 79k | 1k | 114 | 103 |
| 81 | Sep 21, '23 | Woke up to 17k followers Today π A total of 5000 new souls that decided to join my journey since the whole spiral art / optical AI illusion movement started. Thank you everyone fo | 5.2Γ | 79k | 1.1k | 51 | 93 |
| 82 | Jan 8, '25 | adobe just released a method that can generate transparent videos from text and images π | 5.1Γ | 77k | 1.1k | 114 | 535 |
| 83 | Apr 10, '25 | show me your most intricate art. no words. only art. | 4.5Γ | 68k | 1.5k | 89 | 165 |
| 84 | Jan 21, '25 | show me your art. no words. only art. | 4.5Γ | 67k | 925 | 67 | 81 |
| 85 | Sep 19, '23 | GN dreamers β¨ Checkers anyone? | 4.4Γ | 66k | 2k | 138 | 95 |
| 86 | Feb 12, '25 | show me your art. no words. just art. | 4.4Γ | 66k | 945 | 53 | 112 |
| 87 | Jun 21, '25 | feed me your slop. no words. only slop π | 4.3Γ | 64k | 1k | 109 | 161 |
| 88 | Aug 5, '25 | show me your art. no words. only art π | 3.9Γ | 59k | 984 | 91 | 93 |
| 89 | Dec 14, '24 | InstantRestore can restore badly damaged face images with just a few reference images π₯ | 3.9Γ | 59k | 999 | 117 | 711 |
| 90 | Sep 21, '25 | View post on X | 3.8Γ | 56k | 2.3k | 252 | 355 |
| 91 | Jul 28, '25 | show me your art. no words. only art π | 3.7Γ | 55k | 1.4k | 154 | 204 |
| 92 | Sep 20, '23 | GN dreamers β¨ May the triangle guide you on your journeys πΊ | 3.2Γ | 49k | 1.7k | 162 | 57 |
| 93 | May 17, '25 | me and my horse when a new ai model drops | 2.9Γ | 43k | 977 | 97 | 164 |
| 94 | Oct 19, '25 | good morning and a happy sunday | 2.9Γ | 43k | 1.8k | 133 | 235 |
| 95 | Oct 12, '25 | View post on X | 2.4Γ | 35k | 951 | 98 | 132 |
| 96 | Oct 1, '25 | good morning | 2.1Γ | 32k | 1.2k | 105 | 149 |
| 97 | Sep 25 | oh you're my dream girl in a dream world i think it's real girl Dream Girl w/ @gh0stpen π₯ | 1.9Γ | 28k | 138 | 16 | 31 |
| 98 | Oct 31, '22 | The new infinite image feature by @runwayml it pretty cool #aiia #aiiart | 1.4Γ | 0 | 1.3k | 188 | 289 |
| 99 | Sep 29, '22 | Next #deforum music video experiment is rendering πΆπ΅ this time utilizing the piano notes to zoom out. Stay tuned for the entire version. #aiia #aiiart #animation | 0.9Γ | 0 | 920 | 142 | 78 |
@karpathy
@levelsio
@paulg
@naval
@marclou
@illyism