正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Game generation → diffusion output, end-to-end. From prompt to scene: a moon-like environment created in our engine and reskinned into flexible media assets. Built on our web app.

Moonlake

10,278 subscribers

20,014 次观看 • 7 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

We refreshed our logo to better reflect what we are building: multimodal world models that simulate environments, predict outcomes, and enable planning over long horizons. We kept the moon from our original mark and evolved it into a symbol of depth, generation, and evolving worlds.

Moonlake

13,473 次观看 • 5 个月前

Real-time generation isn’t about predicting the next pixel, it requires understanding the structure and context of a 3D world. That’s the shift. Read more about our real-time game-native diffusion model on our blog. Moonlake's Reverie v0.1.0

Real-time generation isn’t about predicting the next pixel, it requires understanding the structure and context of a 3D world. That’s the shift. Read more about our real-time game-native diffusion model on our blog. Moonlake's Reverie v0.1.0

Moonlake

14,439 次观看 • 8 个月前

A Closer Look at Alchemist AI v3.5🔎 In v1, one agent handled everything. You’d enter a prompt like “make a snake game” and get a single block of code—one input, one output. v2 introduced multiple agents with specialized roles: prompt refinement, frontend, backend, and review. It brought more structure, but the process was still linear. Each step produced a single output. With v3, a single agent could handle multiple tasks. It generated HTML, CSS, JavaScript, searched for assets, and more—all in one go, while maintaining full context. Now in v3.5, we combine both models. Multiple agents work in parallel, each capable of producing multiple outputs at once. The prompt compiler adapts based on what you're building. A 3D game? It prioritizes game engines, rendering, and mechanics. A website or 2D app? It shifts focus to relevant frameworks and tools. The tool handler can call several services simultaneously, feeding context-aware data into the code generator for real-time execution. From single-output generation to parallel, adaptive workflows. In the next breakdown, we’ll dive into the architecture behind our proprietary engine in v3.5.

A Closer Look at Alchemist AI v3.5🔎 In v1, one agent handled everything. You’d enter a prompt like “make a snake game” and get a single block of code—one input, one output. v2 introduced multiple agents with specialized roles: prompt refinement, frontend, backend, and review. It brought more structure, but the process was still linear. Each step produced a single output. With v3, a single agent could handle multiple tasks. It generated HTML, CSS, JavaScript, searched for assets, and more—all in one go, while maintaining full context. Now in v3.5, we combine both models. Multiple agents work in parallel, each capable of producing multiple outputs at once. The prompt compiler adapts based on what you're building. A 3D game? It prioritizes game engines, rendering, and mechanics. A website or 2D app? It shifts focus to relevant frameworks and tools. The tool handler can call several services simultaneously, feeding context-aware data into the code generator for real-time execution. From single-output generation to parallel, adaptive workflows. In the next breakdown, we’ll dive into the architecture behind our proprietary engine in v3.5.

ALCHEMIST AI 🔮

12,689 次观看 • 1 年前

Grok 4.5 in Grok Build created an FPS game in under an hour. The prompt was simple. I told it to write a game design document and pull free assets from the web. Then I had it create a TODO.md with implementation phases and run a loop to build out each phase. SpaceXAI and Cursor cooked here. This model is insane.

Grok 4.5 in Grok Build created an FPS game in under an hour. The prompt was simple. I told it to write a game design document and pull free assets from the web. Then I had it create a TODO.md with implementation phases and run a loop to build out each phase. SpaceXAI and Cursor cooked here. This model is insane.

tetsuo

1,810,289 次观看 • 1 个月前

$Following The Amazon Blueprint For Building A Trillion Dollar Tech Company Amazon was amongst the first wave of tech companies to hit a trillion-dollar valuation, alongside Apple and Microsoft. They achieved it by pioneering a novel and ruthlessly-efficient business model. On the front-end they united all key consumer markets with a one-stop-shop marketplace. On the back-end, they built the world’s largest fleet of cloud computing infrastructure to power their marketplace, turning these assets into a multi-billion dollar revenue generation vehicle in their own right via AWS. Our vision at PinLink is to adapt this model to the Web3 sector and create the Amazon of DePIN. Our front-end marketplace will be the first one-stop-shop for all DePIN asset categories, whilst the fractionalization model will give retail-level consumers access to DePIN revenue streams. Meanwhile, on the back-end, we’re aiming to build out the world’s largest fleet of protocol-owned DePIN assets, which we will sell fractional shares of on the marketplace. These assets will serve as the equivalent of Amazon’s AWS arm, powering revenue generation at an infrastructure level, whilst fees from our marketplace fund evermore asset purchases, creating an unrivaled revenue generation flywheel. This is how blue chip businesses are born.$

Following The Amazon Blueprint For Building A Trillion Dollar Tech Company Amazon was amongst the first wave of tech companies to hit a trillion-dollar valuation, alongside Apple and Microsoft. They achieved it by pioneering a novel and ruthlessly-efficient business model. On the front-end they united all key consumer markets with a one-stop-shop marketplace. On the back-end, they built the world’s largest fleet of cloud computing infrastructure to power their marketplace, turning these assets into a multi-billion dollar revenue generation vehicle in their own right via AWS. Our vision at PinLink is to adapt this model to the Web3 sector and create the Amazon of DePIN. Our front-end marketplace will be the first one-stop-shop for all DePIN asset categories, whilst the fractionalization model will give retail-level consumers access to DePIN revenue streams. Meanwhile, on the back-end, we’re aiming to build out the world’s largest fleet of protocol-owned DePIN assets, which we will sell fractional shares of on the marketplace. These assets will serve as the equivalent of Amazon’s AWS arm, powering revenue generation at an infrastructure level, whilst fees from our marketplace fund evermore asset purchases, creating an unrivaled revenue generation flywheel. This is how blue chip businesses are born.

PinLink

51,209 次观看 • 1 年前

Selected as a best paper finalist at #CVPR2026: PixelDiT from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.

Selected as a best paper finalist at #CVPR2026: PixelDiT from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.

NVIDIA AI

28,275 次观看 • 2 个月前

Tripo P1.0 is live. Introducing Smart Mesh in Tripo Studio — generate structured 3D meshes in ~2 seconds. ⚡ ~2s generation 🧠 Smart topology 🛠 Production-ready assets 🌍 Built for game pipelines, real-time rendering, web 3D tools, and scalable content production From prompt to usable mesh — almost instantly. Try it: #3DAI #3DCG #3dart #Tripo

Tripo P1.0 is live. Introducing Smart Mesh in Tripo Studio — generate structured 3D meshes in ~2 seconds. ⚡ ~2s generation 🧠 Smart topology 🛠 Production-ready assets 🌍 Built for game pipelines, real-time rendering, web 3D tools, and scalable content production From prompt to usable mesh — almost instantly. Try it: #3DAI #3DCG #3dart #Tripo

Tripo🔜SIGGRAPH2026

44,600 次观看 • 5 个月前

transforms a simple 3D scene into a realistic render and is smart enough to add reflections on the ground on its own we're not too far away from diffusion game engines

transforms a simple 3D scene into a realistic render and is smart enough to add reflections on the ground on its own we're not too far away from diffusion game engines

Nucleus☕️

59,568 次观看 • 6 个月前

Wemby is a cheat code 🤯 Check out this end-to-end sequence and watch the game live on the NBA App 🔥

Wemby is a cheat code 🤯 Check out this end-to-end sequence and watch the game live on the NBA App 🔥

NBA

320,050 次观看 • 3 年前

A single prompt. A spark of an idea. And just like that, it evolves into a cinematic-level transformation. Built completely with Seedance 2.0 on Yapper, this shot went from a thought to something that feels straight out of a high-end VFX studio in minutes. No crew. No big production. Just AI bringing visuals to life.

A single prompt. A spark of an idea. And just like that, it evolves into a cinematic-level transformation. Built completely with Seedance 2.0 on Yapper, this shot went from a thought to something that feels straight out of a high-end VFX studio in minutes. No crew. No big production. Just AI bringing visuals to life.

Zara

10,281 次观看 • 5 个月前

🚀 Sol-Attn is here! We present a training-free sparse attention method that accelerates video generation while better preserving quality. Sol-Attn unifies dynamic routing, sparse computation, and approximate correction in a single online-softmax pass: • On-the-fly block thresholding for dynamic yet controllable budgets • Proxy-score reuse to approximate unselected blocks Results (vs dense FlashAttention-3): • Wan 2.1-14B: 2.02× end-to-end • HunyuanVideo-13B: 2.12× end-to-end • LTX 2.3: up to 2.4× end-to-end When integrated into Sol-Engine (with kernel fusion + caching): • Wan 2.1-14B: 3.48× end-to-end • HunyuanVideo-13B: 5.08× end-to-end Already available in Sol-Engine. The B200 kernel is still under further optimization. 🎬 Project: 📄 Paper: 🔗 Code:

🚀 Sol-Attn is here! We present a training-free sparse attention method that accelerates video generation while better preserving quality. Sol-Attn unifies dynamic routing, sparse computation, and approximate correction in a single online-softmax pass: • On-the-fly block thresholding for dynamic yet controllable budgets • Proxy-score reuse to approximate unselected blocks Results (vs dense FlashAttention-3): • Wan 2.1-14B: 2.02× end-to-end • HunyuanVideo-13B: 2.12× end-to-end • LTX 2.3: up to 2.4× end-to-end When integrated into Sol-Engine (with kernel fusion + caching): • Wan 2.1-14B: 3.48× end-to-end • HunyuanVideo-13B: 5.08× end-to-end Already available in Sol-Engine. The B200 kernel is still under further optimization. 🎬 Project: 📄 Paper: 🔗 Code:

Enze Xie

20,752 次观看 • 24 天前

From Tammy Fisch (school.of.fisch) on Instagram: This was a great way to start our day and end the week. I love getting a glimpse into my students’ minds.

From Tammy Fisch (school.of.fisch) on Instagram: This was a great way to start our day and end the week. I love getting a glimpse into my students’ minds.

edutopia

35,658 次观看 • 2 年前

we're expanding our marketing team at OpenSea and looking for a Social Media Lead. this person would have an entrepreneural charter and strategic ownership over our social channels. the right candidate needs to have deep cultural experience within our space and the ability to deliver end-to-end outcomes, from concept through execution and measurement. know someone? are someone? apply below ⬇️

we're expanding our marketing team at OpenSea and looking for a Social Media Lead. this person would have an entrepreneural charter and strategic ownership over our social channels. the right candidate needs to have deep cultural experience within our space and the ability to deliver end-to-end outcomes, from concept through execution and measurement. know someone? are someone? apply below ⬇️

Adam Hollander

156,548 次观看 • 8 个月前

Great games need great environments. We're working on a scene builder takes a prompt or image and generates a cohesive scene to build your game on. LLMs notoriously struggle with spatial reasoning, we've worked hard to fix that. All these scenes were created with a prompt!

Great games need great environments. We're working on a scene builder takes a prompt or image and generates a cohesive scene to build your game on. LLMs notoriously struggle with spatial reasoning, we've worked hard to fix that. All these scenes were created with a prompt!

siddharth ahuja

18,491 次观看 • 6 个月前

1/ Happy to share UniDisc - Unified Multimodal Discrete Diffusion – We train a 1.5 billion parameter transformer model from scratch on 250 million image/caption pairs using a **discrete diffusion objective**. Our model has all the benefits of diffusion models but now in multimodal space! - flexible compute-quality tradeoff, zero-shot inpainting and editing, better control via classifier-free guidance and lower latency! We open source everything - our code, weights and the training dataset.

1/ Happy to share UniDisc - Unified Multimodal Discrete Diffusion – We train a 1.5 billion parameter transformer model from scratch on 250 million image/caption pairs using a discrete diffusion objective. Our model has all the benefits of diffusion models but now in multimodal space! - flexible compute-quality tradeoff, zero-shot inpainting and editing, better control via classifier-free guidance and lower latency! We open source everything - our code, weights and the training dataset.

Mihir Prabhudesai

104,934 次观看 • 1 年前

We’ve improved our Supabase integration – now you can build apps with backend functionality right from your first prompt in Lovable. 1. Connect a Supabase project to Lovable 2. Describe the app you want to build 3. Watch your app get built

We’ve improved our Supabase integration – now you can build apps with backend functionality right from your first prompt in Lovable. 1. Connect a Supabase project to Lovable 2. Describe the app you want to build 3. Watch your app get built

Lovable

56,673 次观看 • 1 年前

Say hello to the Deneb engine. Our team continues to check off milestones and make rapid progress in development. We are now into mainstage engine testing and are rapidly iterating on hardware and software toward a flight-ready system. If you want to build bad a** engines – apply on our website.

Say hello to the Deneb engine. Our team continues to check off milestones and make rapid progress in development. We are now into mainstage engine testing and are rapidly iterating on hardware and software toward a flight-ready system. If you want to build bad a** engines – apply on our website.

Impulse Space

41,433 次观看 • 4 个月前

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion TL;DR: Create 3/4DGS from Video Diffusion Note: Some first inference code released (not all yet). Contributions (cited): • We present DimensionX, a novel framework for generating photorealistic 3D and 4D scenes from only a single image using controllable video diffusion. • We propose ST-Director, which decouples the spatial and temporal priors in video diffusion models by learning (spatial and temporal) dimension-aware modules with our curated datasets. We further enhance the hybriddimension control with a training-free composition approach according to the essence of video diffusion denoising process. • To bridge the gap between video diffusion and real-world scenes, we design a trajectory-aware mechanism for 3D generation and an identity-preserving denoising approach for 4D generation, enabling more realistic and controllable scene synthesis. • Extensive experiments manifest that our DimensionX delivers superior performance in video, 3D, and 4D generation compared with baseline methods.

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion TL;DR: Create 3/4DGS from Video Diffusion Note: Some first inference code released (not all yet). Contributions (cited): • We present DimensionX, a novel framework for generating photorealistic 3D and 4D scenes from only a single image using controllable video diffusion. • We propose ST-Director, which decouples the spatial and temporal priors in video diffusion models by learning (spatial and temporal) dimension-aware modules with our curated datasets. We further enhance the hybriddimension control with a training-free composition approach according to the essence of video diffusion denoising process. • To bridge the gap between video diffusion and real-world scenes, we design a trajectory-aware mechanism for 3D generation and an identity-preserving denoising approach for 4D generation, enabling more realistic and controllable scene synthesis. • Extensive experiments manifest that our DimensionX delivers superior performance in video, 3D, and 4D generation compared with baseline methods.

MrNeRF

17,062 次观看 • 1 年前

We’ve just published our Animal Welfare Strategy – the biggest animal welfare reforms in a generation. We're putting an end to puppy smuggling, consulting on a ban on electric shock collars, and phasing out colony cages for hens.

We’ve just published our Animal Welfare Strategy – the biggest animal welfare reforms in a generation. We're putting an end to puppy smuggling, consulting on a ban on electric shock collars, and phasing out colony cages for hens.

Defra UK

17,701 次观看 • 8 个月前