Loading video...

Video Failed to Load

Go Home

📢 Cascadeur 2025.3 is live! Big update with AI interpolation, Filament renderer, quadruped support, and improved physics tools. — Inbetweening is now AI interpolation — Filament (experimental, Windows): HDRI, lighting, materials — Quadrupeds (alpha): Quick Rigging + AutoPosing (Pro only) — Physics: Angle Constraints in Ragdoll + improved Point Constraint

cascadeur3d's profile picture

Cascadeur – The Future of Animation

13,557 subscribers

158,328 views • 8 months ago •via X (Twitter)

EducationScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

📢 New Cascadeur update is live! We’ve added Inbetweening to auto-generate motion between poses (walk, run, fall & more), upgraded AutoPosing for smoother control and full pelvis rotation, and introduced multi-character Ragdoll with custom collision support. #Cascadeur #3D #3danimation #AI #gamedev #indiedev
3:17

📢 New Cascadeur update is live! We’ve added Inbetweening to auto-generate motion between poses (walk, run, fall & more), upgraded AutoPosing for smoother control and full pelvis rotation, and introduced multi-character Ragdoll with custom collision support. #Cascadeur #3D #3danimation #AI #gamedev #indiedev

Cascadeur – The Future of Animation

53,379 views • 1 year ago

Why the character movement in my custom game engine felt janky and how I fixed it. In a game engine, most often, a character moves using the physics engine. Meaning, the player is not just a coordinate in space but a physical body. It has velocity, it handles collisions, and it interacts with the world. Now, as you might know, physics engines need stability. If you run them at variable framerates, things start breaking. Objects phase through walls or fly off into space because the math becomes unpredictable. This is why most game engines lock their physics loop to a 60Hz fixed rate. But here’s the problem: If you have a high-end system, you don't want to limit it at 60 FPS. That's a waste of good hardware. Now, that said, if the GPU is rendering at 144 FPS but the player's position (physics driven) only updates 60 times a second, it creates a micro-stutter that ruins the "smooth" feel of the game. A good way to fix this is to treat the character as two separate things: 1. The Physics Body (Invisible part): This is the "real" character. It lives in the 60Hz physics world, it moves the player and handles collisions. 2. The Visual Model and Camera (Visible part): This is what the player actually sees. It doesn't care about collisions, its only job is to look nice and smooth at whatever framerate the GPU is pushing. Once you have this separation, you can use interpolation to keep them in sync. Every time the physics clock ticks, you save the previous position of the invisible body before moving it to the new one. Between those ticks, calculate how far we are between the last physics update and the next one. By using this to drive the visible parts of the game, the stutters disappear. The physics loop stays fixed behind the scenes, while the visuals slide smoothly between the snapshots. Example: - Right after a tick: blend_weight= 0.0 (The visual model stays at the old physics position). - Halfway to the next: blend_weight= 0.5 (The visual model slides to the middle point). - Just before the next: blend_weight= 0.9 (The visual model is almost at the new physics position). Pro-Tip A critical mistake I made initially, and one many devs make, is parenting the camera and visible parts directly to the player body. If you do this, the camera inherits the discrete 60Hz physics movement by default. In that setup, interpolation won't work because the camera is "stuck" to the physics clock. For this fix to work you must decouple the camera and visuals from the body and move them separately. Player movement processing in Detis Engine: - fixed_process: Physics runs at 60Hz. Handles collisions and raw movement. - process: Variable rate. Mainly used for player input caching in the player case. - late_process: Variable rate. Handles interpolated camera movement after physics and everything else is done being processed. - render. Submits the final interpolated transforms to the GPU. The test environment in the video is running on an old 2070-based laptop. Hopefully the video compression won't introduce any stutter... I’m sharing this in hopes it helps a fellow dev. Cheers.
0:23

Why the character movement in my custom game engine felt janky and how I fixed it. In a game engine, most often, a character moves using the physics engine. Meaning, the player is not just a coordinate in space but a physical body. It has velocity, it handles collisions, and it interacts with the world. Now, as you might know, physics engines need stability. If you run them at variable framerates, things start breaking. Objects phase through walls or fly off into space because the math becomes unpredictable. This is why most game engines lock their physics loop to a 60Hz fixed rate. But here’s the problem: If you have a high-end system, you don't want to limit it at 60 FPS. That's a waste of good hardware. Now, that said, if the GPU is rendering at 144 FPS but the player's position (physics driven) only updates 60 times a second, it creates a micro-stutter that ruins the "smooth" feel of the game. A good way to fix this is to treat the character as two separate things: 1. The Physics Body (Invisible part): This is the "real" character. It lives in the 60Hz physics world, it moves the player and handles collisions. 2. The Visual Model and Camera (Visible part): This is what the player actually sees. It doesn't care about collisions, its only job is to look nice and smooth at whatever framerate the GPU is pushing. Once you have this separation, you can use interpolation to keep them in sync. Every time the physics clock ticks, you save the previous position of the invisible body before moving it to the new one. Between those ticks, calculate how far we are between the last physics update and the next one. By using this to drive the visible parts of the game, the stutters disappear. The physics loop stays fixed behind the scenes, while the visuals slide smoothly between the snapshots. Example: - Right after a tick: blend_weight= 0.0 (The visual model stays at the old physics position). - Halfway to the next: blend_weight= 0.5 (The visual model slides to the middle point). - Just before the next: blend_weight= 0.9 (The visual model is almost at the new physics position). Pro-Tip A critical mistake I made initially, and one many devs make, is parenting the camera and visible parts directly to the player body. If you do this, the camera inherits the discrete 60Hz physics movement by default. In that setup, interpolation won't work because the camera is "stuck" to the physics clock. For this fix to work you must decouple the camera and visuals from the body and move them separately. Player movement processing in Detis Engine: - fixed_process: Physics runs at 60Hz. Handles collisions and raw movement. - process: Variable rate. Mainly used for player input caching in the player case. - late_process: Variable rate. Handles interpolated camera movement after physics and everything else is done being processed. - render. Submits the final interpolated transforms to the GPU. The test environment in the video is running on an old 2070-based laptop. Hopefully the video compression won't introduce any stutter... I’m sharing this in hopes it helps a fellow dev. Cheers.

Ioannis Koukourakis

48,703 views • 7 months ago

Released SuperCmd v2-1.0.4 🚀 Big fat update this time!! Here's what's new 💫 • Completely local translation supporting 30+ languages, plus a Translate Selection command for translating large texts from any app • Support for custom AI providers via OpenAI and Anthropic-compatible APIs • 10+ search engines for browser searches (Bing, Baidu, Yandex...) and the option to create your own custom search engine • Support for custom Quick AI prompts, each with its own hotkey • Support for Apple Shortcuts with hotkeys • Read Aloud Edge TTS now has a voice catalogue covering 40+ languages and 100+ dialects • Support for selecting the reasoning effort for ChatGPT in AI Chat • Support for backing up SuperCmd data • Added gitignore-style exclusion patterns for the file indexing. • New Commands: Window Fill, Reset Launcher Position, Changelog • More cosmetic changes to the UI and a bunch of bug fixes Other key improvements 🧐 • Better UI for Quick Chat, Next / Prev chat option • Improved file search results, now shows complete dir, complete indexing. • Improved UI for widgets, now takes up the rest of the space in the launcher • Settings now appears in Cmd-Tab while open. This was only about 20%. Read the entire changelog: Many more improvements are coming, with a lot in the queue. Download:
0:45

Released SuperCmd v2-1.0.4 🚀 Big fat update this time!! Here's what's new 💫 • Completely local translation supporting 30+ languages, plus a Translate Selection command for translating large texts from any app • Support for custom AI providers via OpenAI and Anthropic-compatible APIs • 10+ search engines for browser searches (Bing, Baidu, Yandex...) and the option to create your own custom search engine • Support for custom Quick AI prompts, each with its own hotkey • Support for Apple Shortcuts with hotkeys • Read Aloud Edge TTS now has a voice catalogue covering 40+ languages and 100+ dialects • Support for selecting the reasoning effort for ChatGPT in AI Chat • Support for backing up SuperCmd data • Added gitignore-style exclusion patterns for the file indexing. • New Commands: Window Fill, Reset Launcher Position, Changelog • More cosmetic changes to the UI and a bunch of bug fixes Other key improvements 🧐 • Better UI for Quick Chat, Next / Prev chat option • Improved file search results, now shows complete dir, complete indexing. • Improved UI for widgets, now takes up the rest of the space in the launcher • Settings now appears in Cmd-Tab while open. This was only about 20%. Read the entire changelog: Many more improvements are coming, with a lot in the queue. Download:

Shobhit - Building SuperCmd

12,022 views • 17 days ago

Elon Musk just told the world’s richest investors in Riyadh something wild: “Money will eventually become irrelevant.” On stage at the Future Investment Initiative in Saudi Arabia, Elon explained: With relentless AI + robotics progress, we’re heading to a future of such insane abundance that currency itself stops mattering. Only physics will still limit us — energy, electricity, raw materials. Exact words from Elon: “There will still be constraints on power… but at some point, currency becomes irrelevant.” This is the guy shipping Optimus robots and running the planet’s biggest AI cluster telling sovereign wealth funds that cash is about to lose its meaning. Mind officially blown. Watch the 32-second clip and tell me: Do you think we’ll actually live in a world where money no longer matters?
0:32

Elon Musk just told the world’s richest investors in Riyadh something wild: “Money will eventually become irrelevant.” On stage at the Future Investment Initiative in Saudi Arabia, Elon explained: With relentless AI + robotics progress, we’re heading to a future of such insane abundance that currency itself stops mattering. Only physics will still limit us — energy, electricity, raw materials. Exact words from Elon: “There will still be constraints on power… but at some point, currency becomes irrelevant.” This is the guy shipping Optimus robots and running the planet’s biggest AI cluster telling sovereign wealth funds that cash is about to lose its meaning. Mind officially blown. Watch the 32-second clip and tell me: Do you think we’ll actually live in a world where money no longer matters?

Camus

11,748 views • 9 months ago

Big news! Our AI browser for SEO is now available for free to everyone on MacOS and Windows. SERP Lens is out of beta and is the go-to workspace for SEOs and their AI agents. Purpose-built for the work of search, the application runs on live web data and is ready for what comes next. Here are some of the reasons why you need to test it out now: • Built in agentic functions • Select your favorite frontier AI model • Utilize skills trained by top consultants • Connect any MCP server & GSC data • Beautifully designed interface For a short time only, you can test out the AI with complimentary credits within the “Pro” trial; then it's $39 p/mo for continued access (can use the browser + some VPN locations for free). Our dev team, led by Sam Underwood 🇬🇧, has done some incredible work here, so make sure to check it out now and let us know what you think! Take a look for yourself:
0:47

Big news! Our AI browser for SEO is now available for free to everyone on MacOS and Windows. SERP Lens is out of beta and is the go-to workspace for SEOs and their AI agents. Purpose-built for the work of search, the application runs on live web data and is ready for what comes next. Here are some of the reasons why you need to test it out now: • Built in agentic functions • Select your favorite frontier AI model • Utilize skills trained by top consultants • Connect any MCP server & GSC data • Beautifully designed interface For a short time only, you can test out the AI with complimentary credits within the “Pro” trial; then it's $39 p/mo for continued access (can use the browser + some VPN locations for free). Our dev team, led by Sam Underwood 🇬🇧, has done some incredible work here, so make sure to check it out now and let us know what you think! Take a look for yourself:

Brodie Clark

125,361 views • 6 days ago

Beta Blocker just got a big update on Windows and Android. Buckle in 🐾 🎨 New Censor Styles Windows gets a full new lineup of effects: Static, Glitch, RGB Shift, Terminal and lot's more. Android gets also gets most of them too! 🎭 Preset Styles Windows now has themed preset styles, letting you swap between complete censor looks instantly instead of tuning every setting by hand. 🔄 Reverse Censor Reverse Censor is now available on Android, with a new Reverse Strength slider to choose your strength. There’s also a setup popup when enabling it, plus stronger One App coverage through Android Accessibility. On Windows, Reverse Censor is now more consistent across live mode, exports, recording, and virtual cam. 🎯 Per-Detection Overrides Windows now supports per-detection customization, so different detections can have their own style, text, and image instead of all sharing one global setup. 🧹 Fixes & Polish Fixed heavy cursor flickering during active blocking, improved performance for animated effects and larger boxes, and cleaned up several parts of the UI. This update also brings better language access, smoother scrolling, cleaner style controls, updated Android device targeting, support for the new Android settings in packs, cleaner overlay behaviour, and a smoother install/update experience. Oh, and the pack creator program was also updated to support all of the new features, and got a major cleanup, so this is essentially 3 programs getting major updates at the same time 🤗 Enjoy 💕
0:28

Beta Blocker just got a big update on Windows and Android. Buckle in 🐾 🎨 New Censor Styles Windows gets a full new lineup of effects: Static, Glitch, RGB Shift, Terminal and lot's more. Android gets also gets most of them too! 🎭 Preset Styles Windows now has themed preset styles, letting you swap between complete censor looks instantly instead of tuning every setting by hand. 🔄 Reverse Censor Reverse Censor is now available on Android, with a new Reverse Strength slider to choose your strength. There’s also a setup popup when enabling it, plus stronger One App coverage through Android Accessibility. On Windows, Reverse Censor is now more consistent across live mode, exports, recording, and virtual cam. 🎯 Per-Detection Overrides Windows now supports per-detection customization, so different detections can have their own style, text, and image instead of all sharing one global setup. 🧹 Fixes & Polish Fixed heavy cursor flickering during active blocking, improved performance for animated effects and larger boxes, and cleaned up several parts of the UI. This update also brings better language access, smoother scrolling, cleaner style controls, updated Android device targeting, support for the new Android settings in packs, cleaner overlay behaviour, and a smoother install/update experience. Oh, and the pack creator program was also updated to support all of the new features, and got a major cleanup, so this is essentially 3 programs getting major updates at the same time 🤗 Enjoy 💕

Isla

76,744 views • 2 months ago

The wait is over 365 days ago, we launched our closed Alpha Since then, we have: ✅ Developed a custom ReAct agent with improved prompting, skill handling, and Nillion data storage ✅ Revealed some of our partnerships (big ones have yet to be announced 👀) ✅ Migrated to a new Uniswap Base (V3) pool for better price stability ✅ Allowed users to create their own Characters ✅ Added new Skills (so many more to go live) ✅ Overhauled the agent UI/UX ✅ Integrated with Telegram We are not building one single agent that does XYZ. We are building: ➡️ An Agent Framework ➡️ Secure Credential Storage ➡️ Enterprise-grade Access Control ➡️ Request tracking, and a lot of other stuff.. that allows thousands of people to create AI agents under minutes. The greatest leverage is not in using tools It’s in creating the tools that create tools The closed Alpha was our proof of concept What comes next is our proof of vision Full platform launch on February 25th 🌐
0:38

The wait is over 365 days ago, we launched our closed Alpha Since then, we have: ✅ Developed a custom ReAct agent with improved prompting, skill handling, and Nillion data storage ✅ Revealed some of our partnerships (big ones have yet to be announced 👀) ✅ Migrated to a new Uniswap Base (V3) pool for better price stability ✅ Allowed users to create their own Characters ✅ Added new Skills (so many more to go live) ✅ Overhauled the agent UI/UX ✅ Integrated with Telegram We are not building one single agent that does XYZ. We are building: ➡️ An Agent Framework ➡️ Secure Credential Storage ➡️ Enterprise-grade Access Control ➡️ Request tracking, and a lot of other stuff.. that allows thousands of people to create AI agents under minutes. The greatest leverage is not in using tools It’s in creating the tools that create tools The closed Alpha was our proof of concept What comes next is our proof of vision Full platform launch on February 25th 🌐

Skillful AI

25,999 views • 1 year ago

Spectre AI On-Chain Search Engine is LIVE! 📢 We’re excited to announce that the Spectre AI On-Chain Search Engine is now LIVE for the public and holders at: Link can also be found pinned in the main Telegram or Spectre AI Website. Designed to be accessible for everyone, the app offers free core features with advanced Pro functionalities tiered for holders with a minimum of 1000 $spectre. Subscriptions will also be available soon for added flexibility. Spectre AI Utilities Overview: - Landing Page: Featuring BTC and Altcoin charts, Fear & Greed Index, Spectre Trending, X Trending, Partners in Focus, News, and UI Watchlists. - UI Favorites: Multiple chart views tailored to track your preferred projects. - Research Zone: Advanced tools for sentiment analysis, sentiment analysis charts, and Technical Analysis sections. - Sentiment Analysis Pro: AI-driven insights and unique scoring system to gauge market sentiment. - Technical Analysis Mode: Supply and demand zones powered by AI to guide trading strategies. - UI Heatmaps: Visualize top sectors and asset performance across the market. - User Panel: A personalized hub to manage account. - UI News: Indexed Websites and X Tweets. Note: X Bubbles (visual Twitter mapping) and Monarch (real-time AI chatzone) will be deployed post-launch. Mobile features will be upgraded in upcoming releases, with the best experience currently on desktop. What’s Launching Today: The beta of our Search Engine functionalities go live today, setting the stage for an exciting journey. Looking Ahead: This launch marks only the beginning. Our roadmap includes future features like staking, revenue sharing, and more. In the first few days, we’ll focus on server and user management to ensure seamless performance with increased traffic. Important Security Reminder: As a security reminder: there are no airdrops and no downloads— Spectre AI is accessible directly as a web app on both desktop and mobile. Try and connect with a fresh wallet and only the minimum required tokens. Enjoy this video showcase, and enjoy this milestone day with us. Spectre AI is just getting started - stay tuned for much more to come! #ai $SPECT
1:47

Spectre AI On-Chain Search Engine is LIVE! 📢 We’re excited to announce that the Spectre AI On-Chain Search Engine is now LIVE for the public and holders at: Link can also be found pinned in the main Telegram or Spectre AI Website. Designed to be accessible for everyone, the app offers free core features with advanced Pro functionalities tiered for holders with a minimum of 1000 $spectre. Subscriptions will also be available soon for added flexibility. Spectre AI Utilities Overview: - Landing Page: Featuring BTC and Altcoin charts, Fear & Greed Index, Spectre Trending, X Trending, Partners in Focus, News, and UI Watchlists. - UI Favorites: Multiple chart views tailored to track your preferred projects. - Research Zone: Advanced tools for sentiment analysis, sentiment analysis charts, and Technical Analysis sections. - Sentiment Analysis Pro: AI-driven insights and unique scoring system to gauge market sentiment. - Technical Analysis Mode: Supply and demand zones powered by AI to guide trading strategies. - UI Heatmaps: Visualize top sectors and asset performance across the market. - User Panel: A personalized hub to manage account. - UI News: Indexed Websites and X Tweets. Note: X Bubbles (visual Twitter mapping) and Monarch (real-time AI chatzone) will be deployed post-launch. Mobile features will be upgraded in upcoming releases, with the best experience currently on desktop. What’s Launching Today: The beta of our Search Engine functionalities go live today, setting the stage for an exciting journey. Looking Ahead: This launch marks only the beginning. Our roadmap includes future features like staking, revenue sharing, and more. In the first few days, we’ll focus on server and user management to ensure seamless performance with increased traffic. Important Security Reminder: As a security reminder: there are no airdrops and no downloads— Spectre AI is accessible directly as a web app on both desktop and mobile. Try and connect with a fresh wallet and only the minimum required tokens. Enjoy this video showcase, and enjoy this milestone day with us. Spectre AI is just getting started - stay tuned for much more to come! #ai $SPECT

SPECTRE AI

280,739 views • 1 year ago

Seedance 2.5 isn’t just an update—it’s the end of the “reroll era.” We’ve gone from burning credits on N attempts to landing usable shots in 1-2 tries. That’s not just efficiency; it’s a fundamental shift from gambling on pixels to directing physics. It clears three new thresholds: Realism: No more “AI sheen.” It’s just real. Cinematography: Lighting & movement now carry a pro photographer’s intent. Narrative: Localized editing turns clips into directed scenes. I’ve been building end-to-end pipelines on BytePlus Lumina, weaving Seedance Enterprise motion with Seedream keyframes & Seedaudio soundscapes. Seedance 2.5 API is about to launch on BytePlus, and now you can try it directly on Lumina.
0:24

Seedance 2.5 isn’t just an update—it’s the end of the “reroll era.” We’ve gone from burning credits on N attempts to landing usable shots in 1-2 tries. That’s not just efficiency; it’s a fundamental shift from gambling on pixels to directing physics. It clears three new thresholds: Realism: No more “AI sheen.” It’s just real. Cinematography: Lighting & movement now carry a pro photographer’s intent. Narrative: Localized editing turns clips into directed scenes. I’ve been building end-to-end pipelines on BytePlus Lumina, weaving Seedance Enterprise motion with Seedream keyframes & Seedaudio soundscapes. Seedance 2.5 API is about to launch on BytePlus, and now you can try it directly on Lumina.

SARAH

127,863 views • 19 days ago

Beta Blocker 3.6 is here and it introduces a BIG new feature. Here's everything that's new: 🖥️ UI Overhaul. The entire interface has been cleaned up to feel more premium, modern, and less boxed-in. 📐 Better Window Scaling. Wider windows now properly use the available space instead of stretching awkwardly or leaving everything floating in the center. 🎨 Live Preview. The Style tab now includes a real-time preview, so you can see exactly what your block will look like before you enable it. ⚙️ Cleaner Detection Settings. Detection settings have been reorganised and simplified so everything is easier to understand. 🧭 Reworked Help & Settings Layout. Important options are now easier to find with a cleaner settings and help structure. now for the big one… 🔄 REVERSE CENSOR REWORK This is the biggest part of the update. Reverse Censor has been completely rebuilt. Instead of blocking detected areas, it now applies your block across the screen while leaving detected parts visible. This also adds: ✨ New Reverse Censor settings tab 🌀 Blur or mosaic modes 🎚️ Strength controls 🖥️ Full monitor selection support ⚡ Live setting updates while blocking is active BUG FIXES & POLISH 🛠️ Fixed Reverse Censor sizing issues on second monitors, including Windows scaling modes like 125% 🖥️ Improved multi-monitor behaviour for effects and Popup Storm 🧹 Cleaned up the Help tab 🎯 Fixed Auto-Detect changing results after switching tabs 📷 Improved camera handling in virtual camera mode 🎮 Reduced cursor flicker while Beta Blocker is active for a smoother gaming experience 🖤 & lots more smaller QoL improvements throughout the app. sfw
0:20

Beta Blocker 3.6 is here and it introduces a BIG new feature. Here's everything that's new: 🖥️ UI Overhaul. The entire interface has been cleaned up to feel more premium, modern, and less boxed-in. 📐 Better Window Scaling. Wider windows now properly use the available space instead of stretching awkwardly or leaving everything floating in the center. 🎨 Live Preview. The Style tab now includes a real-time preview, so you can see exactly what your block will look like before you enable it. ⚙️ Cleaner Detection Settings. Detection settings have been reorganised and simplified so everything is easier to understand. 🧭 Reworked Help & Settings Layout. Important options are now easier to find with a cleaner settings and help structure. now for the big one… 🔄 REVERSE CENSOR REWORK This is the biggest part of the update. Reverse Censor has been completely rebuilt. Instead of blocking detected areas, it now applies your block across the screen while leaving detected parts visible. This also adds: ✨ New Reverse Censor settings tab 🌀 Blur or mosaic modes 🎚️ Strength controls 🖥️ Full monitor selection support ⚡ Live setting updates while blocking is active BUG FIXES & POLISH 🛠️ Fixed Reverse Censor sizing issues on second monitors, including Windows scaling modes like 125% 🖥️ Improved multi-monitor behaviour for effects and Popup Storm 🧹 Cleaned up the Help tab 🎯 Fixed Auto-Detect changing results after switching tabs 📷 Improved camera handling in virtual camera mode 🎮 Reduced cursor flicker while Beta Blocker is active for a smoother gaming experience 🖤 & lots more smaller QoL improvements throughout the app. sfw

Isla

328,894 views • 3 months ago

Our new AI model, SparksMatter, discovered CaMg₂Si₂ - a Ca-filled Mg-Si Zintl silicide - as a thermoelectric built only from stable, non-toxic, earth-abundant elements. Thermoelectrics are solid-state materials that convert heat directly into electricity (and electricity into cooling) with no moving parts, which makes them a key technology for harvesting the vast amounts of waste heat from engines, industry and electronics, and even fusion - but today's best ones rely on scarce or toxic elements like tellurium, lead and bismuth. This is why an earth-abundant, non-toxic candidate matters. Our model's physical reasoning to come up with the design: Mg₂Si is a known earth-abundant thermoelectric but conducts heat too well; a heavy, weakly bound Ca cation in the Mg-Si framework should scatter phonons while keeping a moderate band gap. It generated 100 Ca-Mg-Si crystals with MatterGen, kept the six within 0.05 eV/atom of the convex hull (via MatterSim), and predicted band gaps of 0.44-0.57 eV and bulk moduli of 53-54 GPa (CGCNN). Follow-up lattice dynamics found three CaMg₂Si₂ polymorphs dynamically stable, with lattice thermal conductivity ≈6 W m⁻¹ K⁻¹ at 300 K and ≈2 at 1000 K. The AI proposed a chemical hypothesis first, then developed and applied a separate generative/physics pipeline to test it, and six surviving structures came back with that hypothesized composition. The video replays the reasoning process. New paper out with Alireza Ghafarollahi in npj Computational Materials: SparksMatter, an AI that runs the full in-silico inorganic materials discovery cycle - ideation, planning, computational experimentation, critique and reporting - from a single plain-language query. Why this matters: conventional ML models for materials are typically single-shot predictors or generators. They can predict a property or propose a structure, but they do not organize the next scientific step. Discovery, instead, works as a loop: hypothesize, test, critique, revise. The key advance here is a deep reasoning layer that incorporates physics to decide which scientific tool to use, how to interpret the result, and what to change as next step. How it works: SparksMatter spawns a suite of AI agents - scientists, planners, coders, reviewers and critics - that write and execute code against materials tools: Materials Project retrieval; MatterGen for generative crystal design conditioned on chemistry, band gap or bulk modulus; MatterSim for relaxation and convex-hull stability; CGCNN for property prediction. Adversarial agents check each phase, the system revises its ideas, plans and code from execution results, documents its own limitations, and delivers a scientific report with a validation roadmap spanning DFT, phonons, transport, synthesis and characterization. Two more discovery tasks SparksMatter ran autonomously: 1⃣Soft inorganic semiconductors: generated 112 structures conditioned on low stiffness and narrowed them to 59 candidates absent from the Materials Project after toxicity, stability, electronic, mechanical and database screening - bulk moduli 11-24 GPa, band gaps 0.4-3.9 eV. 2⃣Lead-free perovskites: filtered 154,879 Materials Project entries to 162 Pb-free ABO₃ candidates meeting structural, toxicity, stability and band-gap criteria, including LaAlO₃, BaZrO₃, SrSnO₃, CaTiO₃ and SrTiO₃. Benchmark: the same three tasks were given to frontier reasoning models acting as expert materials scientists with web browsing but without the generation and prediction tools. A blinded LLM evaluator scored every response ten times on relevance, scientific soundness, novelty, and depth and rigor. SparksMatter scored highest in aggregate, with its strongest advantages in novelty and depth & rigor. Its main limitation was scientific soundness because much of the core screening still relied on surrogate models rather than direct first-principles or experimental validation - a gap the system identified, documented, and mapped out how to close. Takeaway: putting generative models, executable code and physics-based simulators inside the reasoning loop lets an AI propose structures outside existing databases, test them, reject weak candidates, and say what evidence is still missing.
1:15

Our new AI model, SparksMatter, discovered CaMg₂Si₂ - a Ca-filled Mg-Si Zintl silicide - as a thermoelectric built only from stable, non-toxic, earth-abundant elements. Thermoelectrics are solid-state materials that convert heat directly into electricity (and electricity into cooling) with no moving parts, which makes them a key technology for harvesting the vast amounts of waste heat from engines, industry and electronics, and even fusion - but today's best ones rely on scarce or toxic elements like tellurium, lead and bismuth. This is why an earth-abundant, non-toxic candidate matters. Our model's physical reasoning to come up with the design: Mg₂Si is a known earth-abundant thermoelectric but conducts heat too well; a heavy, weakly bound Ca cation in the Mg-Si framework should scatter phonons while keeping a moderate band gap. It generated 100 Ca-Mg-Si crystals with MatterGen, kept the six within 0.05 eV/atom of the convex hull (via MatterSim), and predicted band gaps of 0.44-0.57 eV and bulk moduli of 53-54 GPa (CGCNN). Follow-up lattice dynamics found three CaMg₂Si₂ polymorphs dynamically stable, with lattice thermal conductivity ≈6 W m⁻¹ K⁻¹ at 300 K and ≈2 at 1000 K. The AI proposed a chemical hypothesis first, then developed and applied a separate generative/physics pipeline to test it, and six surviving structures came back with that hypothesized composition. The video replays the reasoning process. New paper out with Alireza Ghafarollahi in npj Computational Materials: SparksMatter, an AI that runs the full in-silico inorganic materials discovery cycle - ideation, planning, computational experimentation, critique and reporting - from a single plain-language query. Why this matters: conventional ML models for materials are typically single-shot predictors or generators. They can predict a property or propose a structure, but they do not organize the next scientific step. Discovery, instead, works as a loop: hypothesize, test, critique, revise. The key advance here is a deep reasoning layer that incorporates physics to decide which scientific tool to use, how to interpret the result, and what to change as next step. How it works: SparksMatter spawns a suite of AI agents - scientists, planners, coders, reviewers and critics - that write and execute code against materials tools: Materials Project retrieval; MatterGen for generative crystal design conditioned on chemistry, band gap or bulk modulus; MatterSim for relaxation and convex-hull stability; CGCNN for property prediction. Adversarial agents check each phase, the system revises its ideas, plans and code from execution results, documents its own limitations, and delivers a scientific report with a validation roadmap spanning DFT, phonons, transport, synthesis and characterization. Two more discovery tasks SparksMatter ran autonomously: 1⃣Soft inorganic semiconductors: generated 112 structures conditioned on low stiffness and narrowed them to 59 candidates absent from the Materials Project after toxicity, stability, electronic, mechanical and database screening - bulk moduli 11-24 GPa, band gaps 0.4-3.9 eV. 2⃣Lead-free perovskites: filtered 154,879 Materials Project entries to 162 Pb-free ABO₃ candidates meeting structural, toxicity, stability and band-gap criteria, including LaAlO₃, BaZrO₃, SrSnO₃, CaTiO₃ and SrTiO₃. Benchmark: the same three tasks were given to frontier reasoning models acting as expert materials scientists with web browsing but without the generation and prediction tools. A blinded LLM evaluator scored every response ten times on relevance, scientific soundness, novelty, and depth and rigor. SparksMatter scored highest in aggregate, with its strongest advantages in novelty and depth & rigor. Its main limitation was scientific soundness because much of the core screening still relied on surrogate models rather than direct first-principles or experimental validation - a gap the system identified, documented, and mapped out how to close. Takeaway: putting generative models, executable code and physics-based simulators inside the reasoning loop lets an AI propose structures outside existing databases, test them, reject weak candidates, and say what evidence is still missing.

Markus J. Buehler

40,356 views • 6 days ago

Special thanks to Google DeepMind for inviting me to try out Genie 3. I'm excited to share my thoughts on this early research prototype and also some of my live recordings below: I spent the whole day playing with the system and when it works, it is truly mind blowing🤯. It is the first neural game engine / world model I have tried that generalizes so well and has long term world consistency. Here’s a couple of examples from my live recording and some thoughts on what it means for the future of gaming, robotics, digital experiences and ASI. Where it shines: - Truly general-purpose and quick startup time. Works exceptionally well for gaming environments but also generalizes to other industrial and real-world scenarios. - It learns physics. Although there are systematic failures even for rigid body physics, it was clear to me that it can learn game engine and non-rigid physics without an underlying engine (and in limit learn from game engines via training data). - It works exceptionally well for stylized environments with characters walking around. This will have implications for concept artists, level designers and game devs. - It is way more fun than video models, indicating that there are high retention consumer experiences waiting to be built with this in the future - Photorealistic walk throughs and drone shots work exceptionally well - Global illumination and lighting works surprisingly well - Visual memory is quite powerful and the same objects approximately remain coherent under occlusion and longer time horizons Open Problems: - Physics is still hard and there are obvious failure cases when I tried the classical intuitive physics experiments from psychology (tower of blocks). - Social and multi-agent interactions are tricky to handle. 1vs1 combat games do not work - Long instruction following and simple combinatorial game logic fails (e.g. collect some points / keys etc, go to the door, unlock and so on) - Action space is limited - It is far from being a real game engines and has a long way to go but this is a clear glimpse into the future. The Future: - It is impressive enough for me to have strong conviction that this is going to disrupt the gaming industry. It is super early days and there are a lot of failures but the writing is on the wall. Lots of challenging scientific, engineering and scaling problems to be solved but it is going to happen in the next 5 years. - This is the final piece before we get full AGI and now I think we are well on our way to truly solve it once something like this is scaled up. In many ways it is more ASI than AGI but this is a matter of definitions. The fidelity and generalizability will reach human-level and quickly surpass humans - People are going to combine this with 3D AI and LLMs to build AAA games.
0:57

Special thanks to Google DeepMind for inviting me to try out Genie 3. I'm excited to share my thoughts on this early research prototype and also some of my live recordings below: I spent the whole day playing with the system and when it works, it is truly mind blowing🤯. It is the first neural game engine / world model I have tried that generalizes so well and has long term world consistency. Here’s a couple of examples from my live recording and some thoughts on what it means for the future of gaming, robotics, digital experiences and ASI. Where it shines: - Truly general-purpose and quick startup time. Works exceptionally well for gaming environments but also generalizes to other industrial and real-world scenarios. - It learns physics. Although there are systematic failures even for rigid body physics, it was clear to me that it can learn game engine and non-rigid physics without an underlying engine (and in limit learn from game engines via training data). - It works exceptionally well for stylized environments with characters walking around. This will have implications for concept artists, level designers and game devs. - It is way more fun than video models, indicating that there are high retention consumer experiences waiting to be built with this in the future - Photorealistic walk throughs and drone shots work exceptionally well - Global illumination and lighting works surprisingly well - Visual memory is quite powerful and the same objects approximately remain coherent under occlusion and longer time horizons Open Problems: - Physics is still hard and there are obvious failure cases when I tried the classical intuitive physics experiments from psychology (tower of blocks). - Social and multi-agent interactions are tricky to handle. 1vs1 combat games do not work - Long instruction following and simple combinatorial game logic fails (e.g. collect some points / keys etc, go to the door, unlock and so on) - Action space is limited - It is far from being a real game engines and has a long way to go but this is a clear glimpse into the future. The Future: - It is impressive enough for me to have strong conviction that this is going to disrupt the gaming industry. It is super early days and there are a lot of failures but the writing is on the wall. Lots of challenging scientific, engineering and scaling problems to be solved but it is going to happen in the next 5 years. - This is the final piece before we get full AGI and now I think we are well on our way to truly solve it once something like this is scaled up. In many ways it is more ASI than AGI but this is a matter of definitions. The fidelity and generalizability will reach human-level and quickly surpass humans - People are going to combine this with 3D AI and LLMs to build AAA games.

Tejas Kulkarni

87,960 views • 1 year ago

BARRY 🍯 - Meet Barry, a bear hopelessly addicted to store bought honey. Once content with wild beehives, his taste buds now crave the processed sweetness from supermarket shelves. To get his fix, he repeatedly robs convenience stores. Arrested time and again, he somehow keeps getting released, only to strike again. Miraculously, the charges never stick. I guess being a bear has its perks. My end goal here was to use Kling 3.0 for the entire cinematic sequence, including lipsyncing and dialogue to see how it holds up. The video model has improved a great deal and this was a lot of fun to put together. Enjoy! I love telling all sorts of stories, even silly ones like this. P.S I'm starring in my own short film, I'm the one who is in prison with the bear. Tools For Creation: Image Generation: Nano Banana Pro Image To Video: Kling AI 3.0 Lipsync And Dialogue: Kling 3.0 Music: Elevenlabs & Suno Sound Effects: Artlist Video Editing: Capcut
4:03

BARRY 🍯 - Meet Barry, a bear hopelessly addicted to store bought honey. Once content with wild beehives, his taste buds now crave the processed sweetness from supermarket shelves. To get his fix, he repeatedly robs convenience stores. Arrested time and again, he somehow keeps getting released, only to strike again. Miraculously, the charges never stick. I guess being a bear has its perks. My end goal here was to use Kling 3.0 for the entire cinematic sequence, including lipsyncing and dialogue to see how it holds up. The video model has improved a great deal and this was a lot of fun to put together. Enjoy! I love telling all sorts of stories, even silly ones like this. P.S I'm starring in my own short film, I'm the one who is in prison with the bear. Tools For Creation: Image Generation: Nano Banana Pro Image To Video: Kling AI 3.0 Lipsync And Dialogue: Kling 3.0 Music: Elevenlabs & Suno Sound Effects: Artlist Video Editing: Capcut

Travis Davids

11,125 views • 6 months ago

What if the "flaws" in a system are actually the source code of its intelligence? In new work, we argue that invention behaves like a phase transition driven by exactly this dynamic: novelty is a thermodynamic response to constraint failure. When a system can no longer resolve its inputs within its current degrees of freedom, it is forced to expand its representational space - introducing new effective variables to restore feasibility. Thus innovation is not an accident; it is what a viable system does when the old model stops closing. This allowed us to extract the shared mechanics behind diverse phenomena: rote discovery, creativity, and the spark of insight. We show that symmetry breaking is the new optimization. We exhaustively mapped the topological landscape of matter and musical systems and found that the stabilizing vector is selective imperfection: a specific topological regime that rejects both sterile perfection and incoherent randomness. Strikingly, whether in the Hall-Petch strengthening of high-entropy alloys, function-driving geometry of proteins, or the cultural evolution of musical scales, the corridor for maximum coherence and adaptability is defined by a calculated defect. The physics of resilience and the mathematics of beauty appear to be running the same algorithm. This allows us to hack the vibrational stack by treating vibration as a universal isomorphic operator. We are liquefying the boundary between matter, sound, and intelligence, creating an epistemic inversion: listening becomes a form of seeing and creating. We are translating femtosecond molecular vibrations into audible spectra to design de novo proteins by creating direct lines of communication between Bach and deep-time evolution, and using the "glitch" logic of biology to build swarm AI. The distinction between a spider web’s stress tensor and a musical composition is collapsing; both are generative acts of world-building under constraint. For AI, the implication is straightforward: interpolation is not invention. True structural invention requires systems that can metabolize constraint failure - treating it as the exact point where new degrees of freedom are born. With this machines overcome the old paradigm of simply analyzing the world but are building it. We are operationalizing this via small-world topology. When these new degrees of freedom are born, they don't form a random mess; they snap into global coherence via small-world wiring. We found that this specific connectivity of balancing local motifs with long-range shortcuts is the architectural prerequisite for genuine world-building. Preprint with the full analysis to follow - stay tuned. On to 2026, excited to see what it brings!
4:04

What if the "flaws" in a system are actually the source code of its intelligence? In new work, we argue that invention behaves like a phase transition driven by exactly this dynamic: novelty is a thermodynamic response to constraint failure. When a system can no longer resolve its inputs within its current degrees of freedom, it is forced to expand its representational space - introducing new effective variables to restore feasibility. Thus innovation is not an accident; it is what a viable system does when the old model stops closing. This allowed us to extract the shared mechanics behind diverse phenomena: rote discovery, creativity, and the spark of insight. We show that symmetry breaking is the new optimization. We exhaustively mapped the topological landscape of matter and musical systems and found that the stabilizing vector is selective imperfection: a specific topological regime that rejects both sterile perfection and incoherent randomness. Strikingly, whether in the Hall-Petch strengthening of high-entropy alloys, function-driving geometry of proteins, or the cultural evolution of musical scales, the corridor for maximum coherence and adaptability is defined by a calculated defect. The physics of resilience and the mathematics of beauty appear to be running the same algorithm. This allows us to hack the vibrational stack by treating vibration as a universal isomorphic operator. We are liquefying the boundary between matter, sound, and intelligence, creating an epistemic inversion: listening becomes a form of seeing and creating. We are translating femtosecond molecular vibrations into audible spectra to design de novo proteins by creating direct lines of communication between Bach and deep-time evolution, and using the "glitch" logic of biology to build swarm AI. The distinction between a spider web’s stress tensor and a musical composition is collapsing; both are generative acts of world-building under constraint. For AI, the implication is straightforward: interpolation is not invention. True structural invention requires systems that can metabolize constraint failure - treating it as the exact point where new degrees of freedom are born. With this machines overcome the old paradigm of simply analyzing the world but are building it. We are operationalizing this via small-world topology. When these new degrees of freedom are born, they don't form a random mess; they snap into global coherence via small-world wiring. We found that this specific connectivity of balancing local motifs with long-range shortcuts is the architectural prerequisite for genuine world-building. Preprint with the full analysis to follow - stay tuned. On to 2026, excited to see what it brings!

Markus J. Buehler

67,445 views • 7 months ago

Eric Schmidt just told Congress the number that kills the AI race on Earth: 92 gigawatts of new power, and we can’t deliver it. Former Google CEO laid out math everyone’s ignoring. Average nuclear plant: 1.5 gigawatts. AI demand: 92 gigawatts. That’s 60+ new nuclear facilities needed now, not decades from now. Schmidt: “We need 92 gigawatts more power.” Not happening. Infrastructure doesn’t exist. Approval takes years. Grid physically can’t absorb it. We’re out of electricity. Schmidt investing in Relativity Space isn’t billionaire space hobby. He spotted the bottleneck killing everything and he’s building the only exit that works. Can’t build power plants on Earth fast enough? Move compute off Earth. Schmidt: “You see the problem.” AI doesn’t hit an algorithm wall or chip shortage. It hits power ceiling. The grid can’t deliver 92 gigawatts at the speed AI development demands. Physically impossible to build that capacity terrestrially in relevant timeframes. Not a grid problem. A location problem. Next phase of compute can’t happen on the surface. Period. Heat, power draw, infrastructure limits, all of it forces migration to orbit. Only place with unlimited energy and zero conflicts is space. Schmidt: “We’re running out of electricity.” Direct assessment from someone watching what’s actually being deployed. The gap separating what AI needs and what Earth can provide is unbridgeable at required speeds. Not technical constraints. Physical reality. His aerospace play isn’t exploration. It’s escape route from a grid approaching collapse under computational demand it was never designed to handle. Scaling AI to the levels every major company is planning requires abandoning the planet. Not eventually. Now. Because the alternative is power walls that stop everything regardless of algorithmic genius or hardware breakthroughs. Doesn’t matter how perfect your models are or how many chips you fabricate if you can’t turn them on. And Earth can’t generate power fast enough for what the next five years require. Space isn’t the ambitious choice anymore. It’s the only choice avoiding hard physics limits on how fast you can deploy power generation on a regulated planetary surface. The AI race doesn’t end when someone builds superior intelligence. It ends when they can’t power it while competitors in orbit operate without energy ceilings. And that’s not distant future. That’s the constraint arriving right now that nobody building exclusively on Earth has an answer for.
0:22

Eric Schmidt just told Congress the number that kills the AI race on Earth: 92 gigawatts of new power, and we can’t deliver it. Former Google CEO laid out math everyone’s ignoring. Average nuclear plant: 1.5 gigawatts. AI demand: 92 gigawatts. That’s 60+ new nuclear facilities needed now, not decades from now. Schmidt: “We need 92 gigawatts more power.” Not happening. Infrastructure doesn’t exist. Approval takes years. Grid physically can’t absorb it. We’re out of electricity. Schmidt investing in Relativity Space isn’t billionaire space hobby. He spotted the bottleneck killing everything and he’s building the only exit that works. Can’t build power plants on Earth fast enough? Move compute off Earth. Schmidt: “You see the problem.” AI doesn’t hit an algorithm wall or chip shortage. It hits power ceiling. The grid can’t deliver 92 gigawatts at the speed AI development demands. Physically impossible to build that capacity terrestrially in relevant timeframes. Not a grid problem. A location problem. Next phase of compute can’t happen on the surface. Period. Heat, power draw, infrastructure limits, all of it forces migration to orbit. Only place with unlimited energy and zero conflicts is space. Schmidt: “We’re running out of electricity.” Direct assessment from someone watching what’s actually being deployed. The gap separating what AI needs and what Earth can provide is unbridgeable at required speeds. Not technical constraints. Physical reality. His aerospace play isn’t exploration. It’s escape route from a grid approaching collapse under computational demand it was never designed to handle. Scaling AI to the levels every major company is planning requires abandoning the planet. Not eventually. Now. Because the alternative is power walls that stop everything regardless of algorithmic genius or hardware breakthroughs. Doesn’t matter how perfect your models are or how many chips you fabricate if you can’t turn them on. And Earth can’t generate power fast enough for what the next five years require. Space isn’t the ambitious choice anymore. It’s the only choice avoiding hard physics limits on how fast you can deploy power generation on a regulated planetary surface. The AI race doesn’t end when someone builds superior intelligence. It ends when they can’t power it while competitors in orbit operate without energy ceilings. And that’s not distant future. That’s the constraint arriving right now that nobody building exclusively on Earth has an answer for.

Dustin

160,358 views • 6 months ago

I’d like to introduce you to Thumbnail Academy, the only platform built to help creators design thumbnails that actually get clicks and views Enrollment is open right now I’ve partnered with Dill and ant Together we’ve spent over a decade designing thumbnails for some of the biggest YouTubers on the planet Across billions of views, we’ve learned exactly what makes people click Now we’ve turned everything we know into a system that any creator can use to master thumbnails Until now, creators paid $500+ for a single thumbnail or a one-hour consult We wanted to build something better. Something that helps creators grow from every angle: • Interactive lessons, templates, AI tools, and weekly live coaching calls • Learn how to craft ideas that spark curiosity and pull people in • Master thumbnail psychology and design faster with AI • Understand exactly what drives clicks and how to do it again and again • No fluff. Just the systems behind billions of views. But Thumbnail Academy isn’t just a course It’s also a community built where you’ll connect with other creators, share wins, post your work for feedback, and grow together Inside, you’ll find spaces for learning, inspiration, and collaboration, and for the Pro Tier: live weekly calls where we workshop thumbnails, share AI tools, and help you level up in real time This is the complete blueprint we use every day for the world’s top creators Every lesson, every example, every bit of feedback is built to help you create scroll-stopping thumbnails that actually grow your channel Enrollment closes this Friday After that, the Founders Price is gone for good, and we’ll shift focus to helping the first wave of creators inside If you’ve ever felt frustrated that your video didn’t get the clicks it deserved, this is your chance to fix that for good Join Thumbnail Academy today Your thumbnails, and your channel, will never be the same With love, David (link to Thumbnail Academy on next post!!!)
2:45

I’d like to introduce you to Thumbnail Academy, the only platform built to help creators design thumbnails that actually get clicks and views Enrollment is open right now I’ve partnered with Dill and ant Together we’ve spent over a decade designing thumbnails for some of the biggest YouTubers on the planet Across billions of views, we’ve learned exactly what makes people click Now we’ve turned everything we know into a system that any creator can use to master thumbnails Until now, creators paid $500+ for a single thumbnail or a one-hour consult We wanted to build something better. Something that helps creators grow from every angle: • Interactive lessons, templates, AI tools, and weekly live coaching calls • Learn how to craft ideas that spark curiosity and pull people in • Master thumbnail psychology and design faster with AI • Understand exactly what drives clicks and how to do it again and again • No fluff. Just the systems behind billions of views. But Thumbnail Academy isn’t just a course It’s also a community built where you’ll connect with other creators, share wins, post your work for feedback, and grow together Inside, you’ll find spaces for learning, inspiration, and collaboration, and for the Pro Tier: live weekly calls where we workshop thumbnails, share AI tools, and help you level up in real time This is the complete blueprint we use every day for the world’s top creators Every lesson, every example, every bit of feedback is built to help you create scroll-stopping thumbnails that actually grow your channel Enrollment closes this Friday After that, the Founders Price is gone for good, and we’ll shift focus to helping the first wave of creators inside If you’ve ever felt frustrated that your video didn’t get the clicks it deserved, this is your chance to fix that for good Join Thumbnail Academy today Your thumbnails, and your channel, will never be the same With love, David (link to Thumbnail Academy on next post!!!)

David Altizer

54,513 views • 10 months ago

Test of Vidu 2.0 Update Vidu is the equivalent of Veo 2 for creating Anime with AI! #vidu #viduspeed (Music by Takachan: Hi everyone, hope you're well ^^ 😄👍 It's been a while since I posted because I've been working on an animation. I struggled a lot with character and background consistency issues, so it's still in production! 😉 Today, we are here to test Vidu version 2.0, which will be released in 6 days (By the way, did you spot the Anime references I included? ha ha). Here’s what you need to know: In 3 words: Speed, stability, dynamism ✔ Speed: This new version allows you to animate an image in a few seconds, much faster than previous versions. ✔ Stability: Until now, Vidu was way too dynamic, and we had very little control over it, making storytelling particularly complicated. This version is extremely stable, allowing you to create stories easily. ✔ Image Style Fidelity: This version is extremely faithful to your image styles, something the previous version of Vidu had a lot of trouble with, as it always tried to change the appearance and texture of characters. This version perfectly retains the image style! For example, at 00:19. ✔ Transitions: The transitions have been greatly improved; they are also very stable! At 00:16, you have examples of transitions. It's also a big step forward! ✔ Better Prompt Understanding: Another excellent thing for storytelling! ✔ I also note Vidu 2.0's ability to identify tiny characters and animate them dynamically, which is really great!! You can see it at 00:22. In this animation, I tried to use all of these features. ❌ Negative Point: This exceptional stability gain is at the expense of movement. To continue having very dynamic renders (for example, a character running extremely fast), you'll either need to use keyframes or switch to version 1.5 (which has also been improved). Music by Takachan : ai aiart aianimation aianime ainews anime animenews aitools aitool Animations created with the help of Vidu AI: @Viduforhuman
1:41

Test of Vidu 2.0 Update Vidu is the equivalent of Veo 2 for creating Anime with AI! #vidu #viduspeed (Music by Takachan: Hi everyone, hope you're well ^^ 😄👍 It's been a while since I posted because I've been working on an animation. I struggled a lot with character and background consistency issues, so it's still in production! 😉 Today, we are here to test Vidu version 2.0, which will be released in 6 days (By the way, did you spot the Anime references I included? ha ha). Here’s what you need to know: In 3 words: Speed, stability, dynamism ✔ Speed: This new version allows you to animate an image in a few seconds, much faster than previous versions. ✔ Stability: Until now, Vidu was way too dynamic, and we had very little control over it, making storytelling particularly complicated. This version is extremely stable, allowing you to create stories easily. ✔ Image Style Fidelity: This version is extremely faithful to your image styles, something the previous version of Vidu had a lot of trouble with, as it always tried to change the appearance and texture of characters. This version perfectly retains the image style! For example, at 00:19. ✔ Transitions: The transitions have been greatly improved; they are also very stable! At 00:16, you have examples of transitions. It's also a big step forward! ✔ Better Prompt Understanding: Another excellent thing for storytelling! ✔ I also note Vidu 2.0's ability to identify tiny characters and animate them dynamically, which is really great!! You can see it at 00:22. In this animation, I tried to use all of these features. ❌ Negative Point: This exceptional stability gain is at the expense of movement. To continue having very dynamic renders (for example, a character running extremely fast), you'll either need to use keyframes or switch to version 1.5 (which has also been improved). Music by Takachan : ai aiart aianimation aianime ainews anime animenews aitools aitool Animations created with the help of Vidu AI: @Viduforhuman

Naegiko

20,293 views • 1 year ago