Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

BIG: We are introducing Multi-Scenes🔆, a way to create multiple scenes within the same Spline file + new updates! 🪟 Multi-Scenes ⚡️ Scene Transitions 🏃‍♀️ Animation support on imported models 🎯 Trigger Areas 🌔 Bump & Roughness Maps 💄 UI refresh Demo: Remix:

Spline

111,079 subscribers

82,386 views • 3 years ago •via X (Twitter)

Science & Technology Education

Anya Rossi• Live Now

Private livecam show

11 Comments

Spline3 years ago

Scenes are the equivalent of "Pages" in 2d environments. This means that they can be used for many purposes, like: - Exploring ideas without affecting the main design - Creating multiverse experiences - Organizing components & design systems And more!

RedDeer.Games2 years ago

We can't spill the beans about the release date of Maki: Paw of Fury, but make no mistake, things are happening! 🫘😎 We remind you that the game is coming to #NintendoSwitch and #PC #Steam and you can play the demo on PC, here ⤵️ >>> Have a great day!

Ξugene3 years ago

Wow!! Animation support & bump/roughness you guys are the best! 🙌

Spline3 years ago

You nailed it already! 💕

sambø3 years ago

Multiverse of Spline here we goo 😎

Spline3 years ago

⚡️⚡️⚡️

Guillaume3 years ago

"Bump & Roughness Maps" dingz ✨

Framer University3 years ago

Wtf 😳 It is unbelievable how much Spline has improved since I used it heavily last year.

Marina💫3 years ago

So proud to be a part of this team. It's just wonderful 🔥🥹

Mark Linington3 years ago

This is so cool❤️, do the scenes need to load all at once, or can you load scene by scene progressively?. Ex: you make a menu scene that loads first quickly, as you wait for the next scene to load in background?, so you break up the loading of parts so it’s less long. 🤞

bigloong3 years ago

This is a huge update, it's awesome to finally have bump mapping and to finally support skeletal animation-like effects like walking and running!

Related Videos

Mixboard just got a major tune up! Today, we’re introducing a bundle of updates (with a ribbon on top 🎁) to help you keep exploring your ideas: - Nano Banana Pro: Create presentations with the content found directly on your boards using our latest image generation model - New File Types: Now also supporting PDF, HEIC, & TIFF - Multi-Board Projects: Organize multiple boards within one cohesive project - & more! Try it now ->

Mixboard just got a major tune up! Today, we’re introducing a bundle of updates (with a ribbon on top 🎁) to help you keep exploring your ideas: - Nano Banana Pro: Create presentations with the content found directly on your boards using our latest image generation model - New File Types: Now also supporting PDF, HEIC, & TIFF - Multi-Board Projects: Organize multiple boards within one cohesive project - & more! Try it now ->

Google Labs

455,385 views • 9 months ago

Introducing FlowGPT: The Ultimate AI Video Tool! FlowGPT offers all the features you love in AI video tools,with incredible upgrades: →FlowStudio: From one-liners to 2-minute films 🚀 →AI Script Generation: Automatically create engaging scripts. → Auto Multi-Scene Creation: Seamlessly generate multiple scenes . → Smart Character Voiceovers: Bring characters to life with realistic voices. → One-Click Subtitles & BGM: Effortlessly add subtitles and background music. All this at the same price as Pika, but with the ability to create videos that are 50x longer! 🏆 Prompt: "Naruto uses Talk no Jutsu on Trump" Try it now:

Introducing FlowGPT: The Ultimate AI Video Tool! FlowGPT offers all the features you love in AI video tools,with incredible upgrades: →FlowStudio: From one-liners to 2-minute films 🚀 →AI Script Generation: Automatically create engaging scripts. → Auto Multi-Scene Creation: Seamlessly generate multiple scenes . → Smart Character Voiceovers: Bring characters to life with realistic voices. → One-Click Subtitles & BGM: Effortlessly add subtitles and background music. All this at the same price as Pika, but with the ability to create videos that are 50x longer! 🏆 Prompt: "Naruto uses Talk no Jutsu on Trump" Try it now:

kamran Hassan

10,323 views • 2 years ago

So Seedance 2.0 Omni’s biggest breakthrough might be lip sync and dialogue. It lets you upload audio and sync it to characters, even multiple characters within the same scene. So we ran tests against the leading lip sync tool, HeyGen, and the results were surprising. Seedance Omni held up well, especially in wide cinematic shots and multi-character scenes. The results can still feel a bit unnatural at times, though they improve with iteration. HeyGen continues to perform better for upfront avatar-style lip sync. Still, the fact that we now have a lip sync tool within a video model that can directly compete with HeyGen is real progress.

Curious Refuge

11,660 views • 6 months ago

🚨 SIGGRAPH Asia 2025 Paper Alert 🚨 ➡️Paper Title: WorldExplorer: Towards Generating Fully Navigable 3D Scenes 🌟Few pointers from the paper 🎯Generating 3D worlds from text is a highly anticipated goal in computer vision. Existing works are limited by the degree of exploration they allow inside of a scene, i.e., produce stretched-out and noisy artifacts when moving beyond central or panoramic perspectives. 🎯 To this end, authors of this paper proposed “WorldExplorer”, a novel method based on autoregressive video trajectory generation, which builds fully navigable 3D scenes with consistent visual quality across a wide range of viewpoints. 🎯They initialize their scenes by creating multi-view consistent images corresponding to a 360 degree panorama. 🎯Then, they expanded it by leveraging video diffusion models in an iterative scene generation pipeline. 🎯Concretely, they generated multiple videos along short, pre-defined trajectories, that explore the scene in depth, including motion around objects. 🎯Their novel scene memory conditions each video on the most relevant prior views, while a collision-detection mechanism prevents degenerate results, like moving into objects. 🎯Finally,they fuse all generated views into a unified 3D representation via 3D Gaussian Splatting optimization. 🎯Compared to prior approaches, WorldExplorer produces high-quality scenes that remain stable under large camera motion, enabling for the first time realistic and unrestricted exploration. 🎯They believe this marks a significant step toward generating immersive and truly explorable virtual 3D environments. 🏢Organization: TU München 🧙Paper Authors: Manuel-Andreas Schneider, Lukas Höllein , Matthias Niessner 📝 Read the Full Paper here: 🗂️ Project Page: 🧑‍💻 Code: 🎥 Be sure to watch the attached Technical Summary Video - Sound on 🔊🔊 Find this Valuable 💎 ? ♻️QT and teach your network something new Follow me 👣, naveen manwani , for the latest updates on Tech and AI-related news, insightful research papers, and exciting announcements. #SIGGRAPHAsia2025

🚨 SIGGRAPH Asia 2025 Paper Alert 🚨 ➡️Paper Title: WorldExplorer: Towards Generating Fully Navigable 3D Scenes 🌟Few pointers from the paper 🎯Generating 3D worlds from text is a highly anticipated goal in computer vision. Existing works are limited by the degree of exploration they allow inside of a scene, i.e., produce stretched-out and noisy artifacts when moving beyond central or panoramic perspectives. 🎯 To this end, authors of this paper proposed “WorldExplorer”, a novel method based on autoregressive video trajectory generation, which builds fully navigable 3D scenes with consistent visual quality across a wide range of viewpoints. 🎯They initialize their scenes by creating multi-view consistent images corresponding to a 360 degree panorama. 🎯Then, they expanded it by leveraging video diffusion models in an iterative scene generation pipeline. 🎯Concretely, they generated multiple videos along short, pre-defined trajectories, that explore the scene in depth, including motion around objects. 🎯Their novel scene memory conditions each video on the most relevant prior views, while a collision-detection mechanism prevents degenerate results, like moving into objects. 🎯Finally,they fuse all generated views into a unified 3D representation via 3D Gaussian Splatting optimization. 🎯Compared to prior approaches, WorldExplorer produces high-quality scenes that remain stable under large camera motion, enabling for the first time realistic and unrestricted exploration. 🎯They believe this marks a significant step toward generating immersive and truly explorable virtual 3D environments. 🏢Organization: TU München 🧙Paper Authors: Manuel-Andreas Schneider, Lukas Höllein , Matthias Niessner 📝 Read the Full Paper here: 🗂️ Project Page: 🧑‍💻 Code: 🎥 Be sure to watch the attached Technical Summary Video - Sound on 🔊🔊 Find this Valuable 💎 ? ♻️QT and teach your network something new Follow me 👣, naveen manwani , for the latest updates on Tech and AI-related news, insightful research papers, and exciting announcements. #SIGGRAPHAsia2025

naveen manwani

10,578 views • 11 months ago

Sora 2 consistent characters is absolutely insane 🤯 OpenAI just released the ability to create and save consistent characters in Sora 2. And this changes everything for AI UGC ads. Perfect for DTC brands & agencies who need the same creator across multiple videos in different scenes. Here's why this matters: Before this update, every Sora 2 video gave you a different person. You couldn't build brand consistency or create multi-video campaigns with the same "creator." Now you can: → Generate your ideal AI creator once (holding your product) → Save them as a consistent character in Sora 2 → Tag that character in unlimited video prompts → Same person, different scenes (car, bathroom, outdoors, etc.) → Build entire UGC campaigns with one consistent face Say "bye-bye" to spending $500 on a single UGC video. What you can create: → Multi-video testimonial series with the same person → Before/after campaigns with consistent character → Product demo sequences that feel cohesive → Entire ad creative libraries with your "brand ambassador" I recorded a 12-minute Loom showing the exact step-by-step process for creating consistent characters in Sora 2. Want the complete walkthrough? > Comment "SORA" > Like this post And I'll send the Loom over (must be following so I can DM)

Mike Futia

21,845 views • 10 months ago

New Google Gemma 4 12B claims near-26B performance - we tested both! We ran both models locally on one RTX 4090 and gave each the same task: write a self-contained HTML5 canvas animation with real physics in one file without libraries. Three scenes - a Galton board, two blocks colliding off a wall, and a chaotic triple pendulum Outputs: Gemma 4 26B-A4B: 15 GB VRAM usage, 6.9k tokens, 138 tok/s Gemma 4 12B: 9 GB VRAM usage, 8.9k tokens, 80 tok/s Same Gemma 4 family, but the 26B-A4B won every scene and ran ~1.7x faster - on just 4B active params. The 12B stayed very close though, on almost half the VRAM - which makes it the ideal model for a 16 GB laptop

New Google Gemma 4 12B claims near-26B performance - we tested both! We ran both models locally on one RTX 4090 and gave each the same task: write a self-contained HTML5 canvas animation with real physics in one file without libraries. Three scenes - a Galton board, two blocks colliding off a wall, and a chaotic triple pendulum Outputs: Gemma 4 26B-A4B: 15 GB VRAM usage, 6.9k tokens, 138 tok/s Gemma 4 12B: 9 GB VRAM usage, 8.9k tokens, 80 tok/s Same Gemma 4 family, but the 26B-A4B won every scene and ran ~1.7x faster - on just 4B active params. The 12B stayed very close though, on almost half the VRAM - which makes it the ideal model for a 16 GB laptop

atomic.chat

152,658 views • 3 months ago

Introducing AdFlow — a completely new way to create ads. Built for creative professionals and modern creative teams. Yes, it's a node-based editor. And yes, it's purposefully built for performance ads. → Drag and drop to create 100s of variations → Use any AI model you want → Combine multiple scenes into a linear editor → No switching tools. Ever. We've been using AdFlow internally for the past year. Built tens of thousands of videos for our customers with it. We kept asking ourselves — why are we keeping this to ourselves? Today we're opening AdFlow to everyone. And we're sharing 100 ready-to-use AdFlow templates across different product categories. Comment "Flow" and I'll send them your way. (Must be following so I can DM)

Introducing AdFlow — a completely new way to create ads. Built for creative professionals and modern creative teams. Yes, it's a node-based editor. And yes, it's purposefully built for performance ads. → Drag and drop to create 100s of variations → Use any AI model you want → Combine multiple scenes into a linear editor → No switching tools. Ever. We've been using AdFlow internally for the past year. Built tens of thousands of videos for our customers with it. We kept asking ourselves — why are we keeping this to ourselves? Today we're opening AdFlow to everyone. And we're sharing 100 ready-to-use AdFlow templates across different product categories. Comment "Flow" and I'll send them your way. (Must be following so I can DM)

Yinan Na

346,062 views • 5 months ago

Want to create an avatar from a single image? FlexAvatar is a transformer model that creates full 360°, high-quality, and expressive 3D head avatar from just a single portrait image in minutes. Real-time Demo: FlexAvatar's lightweight architecture allows both animation and rendering in real-time, enabling interactive user experiences. To create a new 3D head avatar, only one image is required, e.g., from a webcam. The final avatar is ready after 2 minutes. Architecture: Under the hood, FlexAvatar adopts a transformer-based encoder-decoder design. The encoder maps the input image onto a latent avatar space, while the decoder produces 3D Gaussian attribute maps by incorporating the animation signal via cross-attention. The model learns all facial animations directly from the data without relying on pre-built 3D face models. This equips the avatars with realistic facial expressions. The internal avatar latent space can be conveniently used to integrate additional observations of a person via fitting. This enables use-cases where more than one image of a person is available, e.g., from a phone scan of the person. We train jointly on 2D monocular videos and multi-view data. However, in monocular videos, the animation signal leaks the target viewpoint, causing the model to produce incomplete 3D heads. We call this phenomenon entanglement of driving signal and target viewpoint. To prevent entanglement, we introduce bias sinks. These are learnable tokens that indicate whether a training sample stems from a monocular or a multi-view dataset. During training, the model learns to produce incomplete 3D heads only when the monocular token is present. During inference, FlexAvatar then always uses the multi-view token for which the model has learned to produce complete 3D heads. This simple design allows to combine the generalizability from monocular data with the quality of multi-view data. FlexAvatar summary: - Input: Single-image, phone scan, or monocular video - Output: Full 360° head avatar - Expressive animations - Real-time rendering and animation - Generalization to any portrait - Create a new avatar in 2 minutes - Use bias sinks to combine 2D and 3D data 🏠 🌍 🎥 Great work by Tobias Kirschstein and Simon Giebenhain!

Matthias Niessner

96,371 views • 9 months ago

glm 5.3 flash is 7.5x cheaper, but 3.4x slower than gemini 3.7 flash Z.ai glm 5.3 flash – shipped aug 26, $0.07/$0.25 per 1m Google DeepMind gemini 3.7 flash – shipped aug 13, $0.38/$1.88 per 1m we put the two models on one job: write one html file that draws an animated 3d scene in the browser. no images, no downloads, and it has to look the same on every load. the setup: three scenes – a glass aquarium in a lit room, the solar system, a night city under a thunderstorm. identical brief word for word, reasoning effort high, 64k output cap. the numbers below are not the whole run. they cover the three scenes we kept – the best one per task from each model, the ones in the video. - total generation time for the three scenes #1 gemini 3.7 flash – 10m 36s #2 glm 5.3 flash – 36m 30s - tokens spent on those three scenes #1 glm 5.3 flash – 110k #2 gemini 3.7 flash – 111k - cost of those three scenes #1 glm 5.3 flash – $0.027 #2 gemini 3.7 flash – $0.202 observations: • glm's first 10 attempts: 7 blank pages. it kept inventing short random helpers and forgetting to define one of them. the fix was one line in the brief: use exactly one random helper, named rand(), and don't invent shorthands next to it. next 12 attempts: 11 alive, 0 crashes. • glm spends 66% of its output on reasoning, gemini 57%. that is the whole speed gap. • gemini's storm came back as a black rectangle in 4 of 6 runs. glm's best storm has a branching bolt, lit rain and wet asphalt – for $0.01. conclusion: same three scenes, same token spend – glm 5.3 flash billed $0.027 and took 36m 30s, gemini 3.7 flash billed $0.202 and took 10m 36s. glm wins gemini on price and made the best storm of the whole run follow thehype. for 24/7 ai news, analysis and breakdowns

glm 5.3 flash is 7.5x cheaper, but 3.4x slower than gemini 3.7 flash Z.ai glm 5.3 flash – shipped aug 26, $0.07/$0.25 per 1m Google DeepMind gemini 3.7 flash – shipped aug 13, $0.38/$1.88 per 1m we put the two models on one job: write one html file that draws an animated 3d scene in the browser. no images, no downloads, and it has to look the same on every load. the setup: three scenes – a glass aquarium in a lit room, the solar system, a night city under a thunderstorm. identical brief word for word, reasoning effort high, 64k output cap. the numbers below are not the whole run. they cover the three scenes we kept – the best one per task from each model, the ones in the video. - total generation time for the three scenes #1 gemini 3.7 flash – 10m 36s #2 glm 5.3 flash – 36m 30s - tokens spent on those three scenes #1 glm 5.3 flash – 110k #2 gemini 3.7 flash – 111k - cost of those three scenes #1 glm 5.3 flash – $0.027 #2 gemini 3.7 flash – $0.202 observations: • glm's first 10 attempts: 7 blank pages. it kept inventing short random helpers and forgetting to define one of them. the fix was one line in the brief: use exactly one random helper, named rand(), and don't invent shorthands next to it. next 12 attempts: 11 alive, 0 crashes. • glm spends 66% of its output on reasoning, gemini 57%. that is the whole speed gap. • gemini's storm came back as a black rectangle in 4 of 6 runs. glm's best storm has a branching bolt, lit rain and wet asphalt – for $0.01. conclusion: same three scenes, same token spend – glm 5.3 flash billed $0.027 and took 36m 30s, gemini 3.7 flash billed $0.202 and took 10m 36s. glm wins gemini on price and made the best storm of the whole run follow thehype. for 24/7 ai news, analysis and breakdowns

thehype.

15,997 views • 25 days ago

How to Turn AI Storyboards Into Cinematic Videos with Seedance 2.0 (Complete Tutorial) In this video you will learn how to transform a single AI storyboard into a fully animated cinematic video using Seedance 2.0 and other powerful AI tools. You will learn the exact workflow for creating consistent AI characters, generating professional storyboard sequences, animating multiple scenes, and producing high quality AI videos in minutes. This tutorial covers storyboard creation, character consistency, AI animation, scene transitions, cinematic storytelling, and the tools used to create polished AI films while saving time and reducing production costs. This information is for both new and existing creators who want to create AI films, animated shorts, commercials, or cinematic videos using the latest AI tools in 2026. Like, Retweet and Bookmark. Follow BIG CHRIS Video Credit: AI Bunny 🤝(

How to Turn AI Storyboards Into Cinematic Videos with Seedance 2.0 (Complete Tutorial) In this video you will learn how to transform a single AI storyboard into a fully animated cinematic video using Seedance 2.0 and other powerful AI tools. You will learn the exact workflow for creating consistent AI characters, generating professional storyboard sequences, animating multiple scenes, and producing high quality AI videos in minutes. This tutorial covers storyboard creation, character consistency, AI animation, scene transitions, cinematic storytelling, and the tools used to create polished AI films while saving time and reducing production costs. This information is for both new and existing creators who want to create AI films, animated shorts, commercials, or cinematic videos using the latest AI tools in 2026. Like, Retweet and Bookmark. Follow BIG CHRIS Video Credit: AI Bunny 🤝(

BIG CHRIS

12,548 views • 2 months ago

We partnered with artists, designers, and builders to create new AI tools that solve real problems in their creative workflows. Here’s what’s new: — Introducing Google Pics in Google Workspace: A brand-new image creation & editing tool. Move and resize objects, add text, and translate just by hovering and clicking — Big updates to صافي النيه😉: 1) You can now create with Gemini Omni Flash in Google Flow 2) Google Flow Agent is a multi-step creative partner that reasons and plans complex tasks with you. 3) Google Flow tools are custom tools you can “vibe code” for animations, video effects, text layering & more — Design live with Stitch by Google: Now, you can use text or voice prompts to edit layouts in real time then export those designs straight to code — More creative control in صافي النيه😉Music: Edit songs section by section, remix the style of full songs, and create music videos with our new Gemini Omni Flash model

Google AI

13,954,152 views • 4 months ago

Open weight Kimi K3 crushed cloud frontier GPT 5.6 at 3D destruction physics! Moonshot AI became the first lab to open-source frontier model weights. We tested them on 8x B300 We gave four models the same task: build three self-contained HTML scenes with real physics Prompts: – A monster truck crushing a row of cars – Two cars jumping a canyon and colliding head-on mid-air – A giant anvil drop test flattening cars one by one Outputs: – Kimi K3 (local): 32.7K tokens, $0 – GPT 5.6: 19.2K tokens, $0.30 – Grok 4.5: 55.2K tokens, $0.45 – GLM 5.2: 66.8K tokens, $0.15 Kimi K3 made the most realistic scenes and won all three against every other model. Its truck rides over the cars and crushes them one by one, and the wrecks stay crushed. The canyon scene shows the gap best: Kimi's cars smash into each other and fall into the gorge, GPT's cars fly past the crash and land with no damage. Grok 4.5 and GLM 5.2 surprised us with the collision animation in the second scene: the impact moment is the most epic of the whole run, bumpers fold, wheels fly off, debris flies everywhere. Their other scenes are nowhere near that level

Open weight Kimi K3 crushed cloud frontier GPT 5.6 at 3D destruction physics! Moonshot AI became the first lab to open-source frontier model weights. We tested them on 8x B300 We gave four models the same task: build three self-contained HTML scenes with real physics Prompts: – A monster truck crushing a row of cars – Two cars jumping a canyon and colliding head-on mid-air – A giant anvil drop test flattening cars one by one Outputs: – Kimi K3 (local): 32.7K tokens, $0 – GPT 5.6: 19.2K tokens, $0.30 – Grok 4.5: 55.2K tokens, $0.45 – GLM 5.2: 66.8K tokens, $0.15 Kimi K3 made the most realistic scenes and won all three against every other model. Its truck rides over the cars and crushes them one by one, and the wrecks stay crushed. The canyon scene shows the gap best: Kimi's cars smash into each other and fall into the gorge, GPT's cars fly past the crash and land with no damage. Grok 4.5 and GLM 5.2 surprised us with the collision animation in the second scene: the impact moment is the most epic of the whole run, bumpers fold, wheels fly off, debris flies everywhere. Their other scenes are nowhere near that level

atomic.chat

150,550 views • 1 month ago

I’m thrilled to announce that we just released GraspGen, a multi-year project we have been cooking at NVIDIA Robotics 🚀 GraspGen: A Diffusion-Based Framework for 6-DOF Grasping Grasping is a foundational challenge in robotics 🤖 — whether for industrial picking or general-purpose humanoids. VLA + real data collection is all the rage now but is expensive and scales poorly for this task. For every new gripper and/or scene, you’ll have to recollect the dataset in this paradigm for the best perf. 💡Key Idea: Since grasping is such a well-defined task in simulation - why can’t we just scale synthetic data generation and train a generative model for grasping? By embracing modularity and standardized grasp formats, we can make this a turnkey technology that works zero-shot for multiple settings. GraspGen is a modular framework for diffusion-based 6-DOF grasp generation that scales across embodiment types, observability conditions, clutter, task complexity. Key Features: ✅ Multi-embodiment support: suction, parallel-jaw, and multi-fingered grippers ✅ Generalization to partial + complete 3D point clouds ✅ Generalization to single-objects + cluttered scenes ✅ Modular design uses other robotics modules and foundation models (SAM2, cuRobo, FoundationStereo, FoundationPose). This allows GraspGen to focus on only one thing - grasp generation ✅ Training recipe: grasp discriminator is trained with On-Generator data from the diffusion model - so that it learns to correct the mistakes (if any) of the diffusion generator ✅ Real-time performance (~20 Hz) before any GPU acceleration; low memory footprint 📊 Results: • SOTA on the FetchBench [Han et al. CoRL 2024] benchmark • Zero-shot sim-to-real transfer on unknown objects and cluttered scenes • Dataset of 53M simulated grasps across 8K objects from Objaverse 📄 arXiv: 🌐 Website: 💻 Code: A huge thank you to everyone involved in this journey — excited to see what the community builds on top of it! Joint work with Clemens Eppner , Balakumar Sundaralingam , Yu-Wei, Jun Yamada Wentao Yuan and other collaborators #robotics #diffusionmodels #physicalAI #simtoreal

I’m thrilled to announce that we just released GraspGen, a multi-year project we have been cooking at NVIDIA Robotics 🚀 GraspGen: A Diffusion-Based Framework for 6-DOF Grasping Grasping is a foundational challenge in robotics 🤖 — whether for industrial picking or general-purpose humanoids. VLA + real data collection is all the rage now but is expensive and scales poorly for this task. For every new gripper and/or scene, you’ll have to recollect the dataset in this paradigm for the best perf. 💡Key Idea: Since grasping is such a well-defined task in simulation - why can’t we just scale synthetic data generation and train a generative model for grasping? By embracing modularity and standardized grasp formats, we can make this a turnkey technology that works zero-shot for multiple settings. GraspGen is a modular framework for diffusion-based 6-DOF grasp generation that scales across embodiment types, observability conditions, clutter, task complexity. Key Features: ✅ Multi-embodiment support: suction, parallel-jaw, and multi-fingered grippers ✅ Generalization to partial + complete 3D point clouds ✅ Generalization to single-objects + cluttered scenes ✅ Modular design uses other robotics modules and foundation models (SAM2, cuRobo, FoundationStereo, FoundationPose). This allows GraspGen to focus on only one thing - grasp generation ✅ Training recipe: grasp discriminator is trained with On-Generator data from the diffusion model - so that it learns to correct the mistakes (if any) of the diffusion generator ✅ Real-time performance (~20 Hz) before any GPU acceleration; low memory footprint 📊 Results: • SOTA on the FetchBench [Han et al. CoRL 2024] benchmark • Zero-shot sim-to-real transfer on unknown objects and cluttered scenes • Dataset of 53M simulated grasps across 8K objects from Objaverse 📄 arXiv: 🌐 Website: 💻 Code: A huge thank you to everyone involved in this journey — excited to see what the community builds on top of it! Joint work with Clemens Eppner , Balakumar Sundaralingam , Yu-Wei, Jun Yamada Wentao Yuan and other collaborators #robotics #diffusionmodels #physicalAI #simtoreal

Adithya Murali

24,347 views • 1 year ago

The time has come, and we need your help! Zipped Up has been in development since December of 2023, and our hardworking team deserves compensation for their hard work thus far, and for future. Introducing ... the ZIPPED UP KICKSTARTER! Launching April 2nd all the way to June 2nd, we are more than ready to get this show on the road! Backers will receive rewards, behind the scenes updates, and much more! If you'd like to see Zipped Up come into fruition sooner rather than later, feel free to share this Kickstarter link with your friends and family! We will be in pre-launch until April 2nd, 2025, meaning it is completely free to sign up for our email list! Thank you so much for all of your support thus far, as the BlipToons team could not be anymore grateful for the love and support we have received in the year of progress we've made so far! #indieanimation #animation #zippedup

The time has come, and we need your help! Zipped Up has been in development since December of 2023, and our hardworking team deserves compensation for their hard work thus far, and for future. Introducing ... the ZIPPED UP KICKSTARTER! Launching April 2nd all the way to June 2nd, we are more than ready to get this show on the road! Backers will receive rewards, behind the scenes updates, and much more! If you'd like to see Zipped Up come into fruition sooner rather than later, feel free to share this Kickstarter link with your friends and family! We will be in pre-launch until April 2nd, 2025, meaning it is completely free to sign up for our email list! Thank you so much for all of your support thus far, as the BlipToons team could not be anymore grateful for the love and support we have received in the year of progress we've made so far! #indieanimation #animation #zippedup

WE CREATE TO INSPIRE💚

160,973 views • 1 year ago

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):

Tencent Hy

126,668 views • 7 months ago

Introducing GGEZ: The Nextjs for ThreeJS Games It's an open source framework which adds all the missing pieces to vibe code better ThreeJS games It has full codex integration, so a $20 ChatGPT sub is enough to build games! Literally "bun run start" and you have the full development environment on localhost 0. GGEZ Runtime - Abstraction layer over physics libraries - Character Controllers - It's just ThreeJS, no magic - Load ggez scenes and animations automatically 1. Trident - a World Editor - Codex World agent - A full editor to build scenes - Including Mesh editing, vertex, edge, face - Terrain sculpting - Physics and Player Controller settings - It's just exporting json files and glb assets, no magic 2. Animation Studio - The best you can find on the web - Codex Animation agent - Build state machines and animation graphs - Multi dimensional blend trees - Clip Editor: Create new animations with codex or edit keyframes - Equipment Editor: Never miss the placement of your rifle anymore - ROOT MOTION SUPPORT 3. GGEZ CLI Yea relax, it works fully headless and you can just create new games with bunx create-ggez new-game But at this point just use vanilla threejs?? Anyways if you are like me and you can't guess with code where objects should be placed and you are fifty prompts deep into figuring out where that box should be placed, this is for you If you are an anti AI game developer who insists that this is slop, then just leave a raging comment below please it's good for the algo 🙏 The whole thing is absolutely experimental and things will break as i move very fast, but I will be building my game with it so i will make sure it becomes stable asap! Link to repo below

Introducing GGEZ: The Nextjs for ThreeJS Games It's an open source framework which adds all the missing pieces to vibe code better ThreeJS games It has full codex integration, so a $20 ChatGPT sub is enough to build games! Literally "bun run start" and you have the full development environment on localhost 0. GGEZ Runtime - Abstraction layer over physics libraries - Character Controllers - It's just ThreeJS, no magic - Load ggez scenes and animations automatically 1. Trident - a World Editor - Codex World agent - A full editor to build scenes - Including Mesh editing, vertex, edge, face - Terrain sculpting - Physics and Player Controller settings - It's just exporting json files and glb assets, no magic 2. Animation Studio - The best you can find on the web - Codex Animation agent - Build state machines and animation graphs - Multi dimensional blend trees - Clip Editor: Create new animations with codex or edit keyframes - Equipment Editor: Never miss the placement of your rifle anymore - ROOT MOTION SUPPORT 3. GGEZ CLI Yea relax, it works fully headless and you can just create new games with bunx create-ggez new-game But at this point just use vanilla threejs?? Anyways if you are like me and you can't guess with code where objects should be placed and you are fifty prompts deep into figuring out where that box should be placed, this is for you If you are an anti AI game developer who insists that this is slop, then just leave a raging comment below please it's good for the algo 🙏 The whole thing is absolutely experimental and things will break as i move very fast, but I will be building my game with it so i will make sure it becomes stable asap! Link to repo below

robot

44,009 views • 5 months ago

Quick little demo of the new $GOD Webstore. ⚡️ Use $GOD directly if you're holding it. Base is the fastest/cheapest for individual transactions, but you can use holdings on ETH or Arb Nova also. You'll get VIP points for this, too, just like any other purchase. ⚡️ Can use ETH or USDC if you're not. ⚡️ The Webstore automatically swaps into $GOD on a DEX in the background, behind-the-scenes. The Webstore supports payments on Base, ETH, and Arbitrum Nova out of the box. And unlike the previous 'Journey to Olympus' events, you'll be able to use your King Of Destiny 👑 balances directly also (USDC or $GOD). Similar flows will support credit card and Apple Pay in the future, but start to converge closer to the app itself in our PWA Marketplace so that Web2 users can easily buy — just like any other premium currency. We know things are a bit disjointed currently as we roll out new infra, but the pieces will start to come together quickly. The future of token access in games is frictionless. ✊⚡️

Quick little demo of the new $GOD Webstore. ⚡️ Use $GOD directly if you're holding it. Base is the fastest/cheapest for individual transactions, but you can use holdings on ETH or Arb Nova also. You'll get VIP points for this, too, just like any other purchase. ⚡️ Can use ETH or USDC if you're not. ⚡️ The Webstore automatically swaps into $GOD on a DEX in the background, behind-the-scenes. The Webstore supports payments on Base, ETH, and Arbitrum Nova out of the box. And unlike the previous 'Journey to Olympus' events, you'll be able to use your King Of Destiny 👑 balances directly also (USDC or $GOD). Similar flows will support credit card and Apple Pay in the future, but start to converge closer to the app itself in our PWA Marketplace so that Web2 users can easily buy — just like any other premium currency. We know things are a bit disjointed currently as we roll out new infra, but the pieces will start to come together quickly. The future of token access in games is frictionless. ✊⚡️

Bry DiSanto 🤙

42,557 views • 1 year ago

Alibaba presents MIMO Controllable Character Video Synthesis with Spatial Decomposed Modeling Character video synthesis aims to produce realistic videos of animatable characters within lifelike scenes. As a fundamental problem in the computer vision and graphics community, 3D works typically require multi-view captures for per-case training, which severely limits their applicability of modeling arbitrary characters in a short time. Recent 2D methods break this limitation via pre-trained diffusion models, but they struggle for pose generality and scene interaction. To this end, we propose MIMO, a novel framework which can not only synthesize character videos with controllable attributes (i.e., character, motion and scene) provided by simple user inputs, but also simultaneously achieve advanced scalability to arbitrary characters, generality to novel 3D motions, and applicability to interactive real-world scenes in a unified framework. The core idea is to encode the 2D video to compact spatial codes, considering the inherent 3D nature of video occurrence. Concretely, we lift the 2D frame pixels into 3D using monocular depth estimators, and decompose the video clip to three spatial components (i.e., main human, underlying scene, and floating occlusion) in hierarchical layers based on the 3D depth. These components are further encoded to canonical identity code, structured motion code and full scene code, which are utilized as control signals of synthesis process. The design of spatial decomposed modeling enables flexible user control, complex motion expression, as well as 3D-aware synthesis for scene interactions. Experimental results demonstrate effectiveness and robustness of the proposed method.

Alibaba presents MIMO Controllable Character Video Synthesis with Spatial Decomposed Modeling Character video synthesis aims to produce realistic videos of animatable characters within lifelike scenes. As a fundamental problem in the computer vision and graphics community, 3D works typically require multi-view captures for per-case training, which severely limits their applicability of modeling arbitrary characters in a short time. Recent 2D methods break this limitation via pre-trained diffusion models, but they struggle for pose generality and scene interaction. To this end, we propose MIMO, a novel framework which can not only synthesize character videos with controllable attributes (i.e., character, motion and scene) provided by simple user inputs, but also simultaneously achieve advanced scalability to arbitrary characters, generality to novel 3D motions, and applicability to interactive real-world scenes in a unified framework. The core idea is to encode the 2D video to compact spatial codes, considering the inherent 3D nature of video occurrence. Concretely, we lift the 2D frame pixels into 3D using monocular depth estimators, and decompose the video clip to three spatial components (i.e., main human, underlying scene, and floating occlusion) in hierarchical layers based on the 3D depth. These components are further encoded to canonical identity code, structured motion code and full scene code, which are utilized as control signals of synthesis process. The design of spatial decomposed modeling enables flexible user control, complex motion expression, as well as 3D-aware synthesis for scene interactions. Experimental results demonstrate effectiveness and robustness of the proposed method.

AK

149,079 views • 2 years ago