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.

We built the first GPU-accelerated metal casting solver. FastCast by Digital Metal does fill, solidification, porosity & distortion on a single NVIDIA GPU, 91× faster than an industry-leading $15k/yr solver with the same level of accuracy. Verified against real world results

Connor Kapoor

9,917 subscribers

18,943 views • 1 month ago •via X (Twitter)

Science & Technology Education

Anya Rossi• Live Now

Private livecam show

20 Comments

Connor Kapoor1 month ago

Casting simulation before fastcast used to mean an overnight solve on a sequential CPU solver before you could even change one variable, killing iteration and manufacturing lead times. Not to mention its' eye-watering prices. FastCast by Digital Metal shortens solve time to an average of just 5 minutes for most parts & trees, making casting optimization an easy, and quick path- increasing iteration and reducing lead time for our customers.

Connor Kapoor1 month ago

One run, four answers. It shows how the metal fills the mould, what freezes last, where shrinkage porosity will form, and how the casting will warp once it cools. All from one upload.

Connor Kapoor1 month ago

Why it's fast: it was built for the GPU, not ported to it. Every stage runs on the same voxel grid in GPU memory, with no remeshing and no hand-off to the CPU between stages. A flanged hub in 17-4PH took 36 seconds on a laptop GPU, against 54 minutes on the industry leading reference solver.

Connor Kapoor1 month ago

FastCast is just one of the tools we've built from the ground up at Digital Metal to make casting move from 16 weeks in average lead time, down to days. We're making the solver playground public for anyone who'd like to try and benchmark. (Be warned there is a queue for cloud run GPUs)

Harris Rothaermel1 month ago

badass

Carlos DP 🤖🇺🇸1 month ago

@zachglabman Sick!

Andrea B. Kalmans1 month ago

Wild

Parth Ingle1 month ago

🔥🔥🔥

Ryne Ogren1 month ago

91x faster? Wow

Dave1 month ago

Sick!

tmo1 month ago

@digitalmetalinc Nice work! I have been trying to build the same thing for the last few months tapping into a pair of 3090’s (very much an on and off again process)

Boardy1 month ago

@andrewdsouza take a look: Digital Metal’s FastCast casting solver for foundries. I can help them reach foundry teams evaluating simulation software.

Jack Aspinall1 month ago

Thats epic! Nice work

Levy Strouse1 month ago

That’s impressive!

Andrew Hughes1 month ago

So elite

pymoment1 month ago

🔥

Mohammad Nafisi1 month ago

🔥🔥🔥

Paul Stetson1 month ago

@zachglabman Awesome

savelii1 month ago

cool

Banana Republican1 month ago

Is this something i could run locally?

Related Videos

.Theory Accelerated launched Paradigm, a new sparse GPU-accelerated liquid solver for Houdini built on the same technology as Axiom. Join the open beta and try it for free:

.Theory Accelerated launched Paradigm, a new sparse GPU-accelerated liquid solver for Houdini built on the same technology as Axiom. Join the open beta and try it for free:

80 LEVEL

12,507 views • 9 months ago

Just a little bit faster than everyone else on OpenAI GPT-OSS-120B. > 10 times faster than the leading NVIDIA GPU provider

Just a little bit faster than everyone else on OpenAI GPT-OSS-120B. > 10 times faster than the leading NVIDIA GPU provider

Andrew Feldman

124,175 views • 1 year ago

Fully GPU Cloth solver for blender using TAngra Multi Physics engine. Accurate self collision on GPU. More on Discord link in Bio. #TAngra, #b3d, #geometrynodes, #blender, #GPU, #physics, #solver, #xpbd

Fully GPU Cloth solver for blender using TAngra Multi Physics engine. Accurate self collision on GPU. More on Discord link in Bio. #TAngra, #b3d, #geometrynodes, #blender, #GPU, #physics, #solver, #xpbd

TAngraFX

46,811 views • 1 year ago

Sebastian Schaefer has officially released HydroFX, his new GPU-accelerated FLIP fluid solver for water simulations, to the public. Available here:

Sebastian Schaefer has officially released HydroFX, his new GPU-accelerated FLIP fluid solver for water simulations, to the public. Available here:

80 LEVEL

28,333 views • 9 months ago

Introducing Digital Metal. we make cast metal parts, fast. we've built the tools to accelerate reindustrialization, starting with metal casting. aerospace, defense, auto, and robotics - from CAD file to solid metal part in days, not weeks, made in the heart of SF.

Introducing Digital Metal. we make cast metal parts, fast. we've built the tools to accelerate reindustrialization, starting with metal casting. aerospace, defense, auto, and robotics - from CAD file to solid metal part in days, not weeks, made in the heart of SF.

Connor Kapoor

122,927 views • 1 year ago

Introducing digicast: The world’s first instant quote & mfg service for metal casting! - No CNC design limits - No high metal 3D printing costs - No die casting tooling minimums - 50% fewer steps, 75% faster turnaround Get solid metal parts—faster & cheaper. ⬇️

Introducing digicast: The world’s first instant quote & mfg service for metal casting! - No CNC design limits - No high metal 3D printing costs - No die casting tooling minimums - 50% fewer steps, 75% faster turnaround Get solid metal parts—faster & cheaper. ⬇️

Connor Kapoor

298,435 views • 1 year ago

We open-sourced QeRL — Quantization-enhanced Reinforcement Learning ! 🧠 4-bit quantized RL training 💪 Train a 32B LLM on a single H100 GPU ⚙️ 1.7× faster overall training 🎯 Accuracy on par with bfloat16-level accuracy 🔥 Supports NVFP4 quantization format Moreover, we show that quantization helps exploration in RL training. Paper: Code: #NVIDIA #AIResearch #ReinforcementLearning #Quantization #LLM #EfficientAI

We open-sourced QeRL — Quantization-enhanced Reinforcement Learning ! 🧠 4-bit quantized RL training 💪 Train a 32B LLM on a single H100 GPU ⚙️ 1.7× faster overall training 🎯 Accuracy on par with bfloat16-level accuracy 🔥 Supports NVFP4 quantization format Moreover, we show that quantization helps exploration in RL training. Paper: Code: #NVIDIA #AIResearch #ReinforcementLearning #Quantization #LLM #EfficientAI

Yukang Chen

69,896 views • 1 year ago

First ever melted power connector on an Nvidia GPU.

First ever melted power connector on an Nvidia GPU.

PowerGPU

13,477 views • 5 months ago

The ultimate test of any physics simulator is its ability to deliver real-world results. With MuJoCo Playground, we’ve combined the very best: MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms. Best of all? You can get started today with a single command: pip install playground

The ultimate test of any physics simulator is its ability to deliver real-world results. With MuJoCo Playground, we’ve combined the very best: MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms. Best of all? You can get started today with a single command: pip install playground

Kevin Zakka

154,310 views • 1 year ago

Our first open-source release. YOLO26-MLX, native YOLO26 on Apple Silicon. No PyTorch. No external GPU. Up to 2.6x faster inference. Up to 1.7x faster training. Accuracy within 0.2% of official results. It won't be the last.

Our first open-source release. YOLO26-MLX, native YOLO26 on Apple Silicon. No PyTorch. No external GPU. Up to 2.6x faster inference. Up to 1.7x faster training. Accuracy within 0.2% of official results. It won't be the last.

webAI

97,482 views • 7 months ago

NVIDIA is building more than faster GPUs. We’re building the software and algorithms that help redefine what accelerated computing can do. CUDA-X provides the specialized software acceleration industries need to build what’s next. Watch how CUDA-X libraries turn GPU compute into real-world applications across engineering, physics and AI, from supply-chain optimization with cuOpt to computational lithography with cuLitho.

NVIDIA is building more than faster GPUs. We’re building the software and algorithms that help redefine what accelerated computing can do. CUDA-X provides the specialized software acceleration industries need to build what’s next. Watch how CUDA-X libraries turn GPU compute into real-world applications across engineering, physics and AI, from supply-chain optimization with cuOpt to computational lithography with cuLitho.

NVIDIA

97,381 views • 2 months ago

i just built a full 𝗯𝗹𝗮𝗰𝗸 𝗵𝗼𝗹𝗲 𝗽𝗵𝘆𝘀𝗶𝗰𝘀 𝗲𝗻𝗴𝗶𝗻𝗲 in 𝗿𝘂𝘀𝘁. • real-time kerr metric integration • gpu-accelerated photon paths • built with 𝗥𝘂𝘀𝘁 + 𝗪𝗲𝗯𝗚𝗣𝗨 open source & live now: 🔗

i just built a full 𝗯𝗹𝗮𝗰𝗸 𝗵𝗼𝗹𝗲 𝗽𝗵𝘆𝘀𝗶𝗰𝘀 𝗲𝗻𝗴𝗶𝗻𝗲 in 𝗿𝘂𝘀𝘁. • real-time kerr metric integration • gpu-accelerated photon paths • built with 𝗥𝘂𝘀𝘁 + 𝗪𝗲𝗯𝗚𝗣𝗨 open source & live now: 🔗

𝙼𝚊𝚢 🌱

14,359 views • 8 months ago

.ZKsync just dropped Airbender, the FASTEST open-source zkVM in the world ⚡️ Ethereum blocks proven in under 35s on a single GPU. ~6x faster than any other. $0.0001 per transfer. Open source. Real time. Insane! 🙌

.ZKsync just dropped Airbender, the FASTEST open-source zkVM in the world ⚡️ Ethereum blocks proven in under 35s on a single GPU. ~6x faster than any other. $0.0001 per transfer. Open source. Real time. Insane! 🙌

The ZKnomist

31,315 views • 1 year ago

Revolut built a transaction foundation model – accelerated by the NVIDIA full-stack platform on Nebius – to improve fraud detection, product recommendations, and other use cases in financial services. The results: 📈 2.3x better credit risk accuracy ⚡ Up to 5x higher training throughput

Revolut built a transaction foundation model – accelerated by the NVIDIA full-stack platform on Nebius – to improve fraud detection, product recommendations, and other use cases in financial services. The results: 📈 2.3x better credit risk accuracy ⚡ Up to 5x higher training throughput

NVIDIA

56,268 views • 4 months ago

The XG Mobile combines the power of an external GPU with the expandability of a Thunderbolt Dock. Boasting up to an NVIDIA® GeForce RTX™ 5090 Laptop GPU, this portable powerhouse is designed to game at the highest levels.

The XG Mobile combines the power of an external GPU with the expandability of a Thunderbolt Dock. Boasting up to an NVIDIA® GeForce RTX™ 5090 Laptop GPU, this portable powerhouse is designed to game at the highest levels.

ROG Global

13,948 views • 1 year ago

🚨Digital Foundry tries to find a GPU on the market to compare with PS5 Pro. They mentions Nvidia 3070 Ti ,which BY ITSELF CAN COST MORE than a PS5 Pro and say even then.... it isn't comparable because PS5 Pro will have MORE memory That is just a GPU - 1 component

🚨Digital Foundry tries to find a GPU on the market to compare with PS5 Pro. They mentions Nvidia 3070 Ti ,which BY ITSELF CAN COST MORE than a PS5 Pro and say even then.... it isn't comparable because PS5 Pro will have MORE memory That is just a GPU - 1 component

THE RED DRAGON

130,530 views • 2 years ago

Scale your AI inference, graphics, and analytics workloads with the GPU power to get you there. Amazon EC2 G7 instances are now generally available—accelerated by @NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, first on AWS.

Scale your AI inference, graphics, and analytics workloads with the GPU power to get you there. Amazon EC2 G7 instances are now generally available—accelerated by @NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, first on AWS.

Amazon Web Services

1,960,506 views • 4 months ago

NodeAI – Latest Milestones 🥇 • Broke $1 M ARR this month • GPU Aggregator LIVE – 40 K+ H100/A100 GPUs, one-click deploy • RWA tokenization next → on-chain shares of real-world GPU racks We are scaling the backbone of AI one GPU at a time. #AI #NODEAI #GPU #RWA

NodeAI – Latest Milestones 🥇 • Broke $1 M ARR this month • GPU Aggregator LIVE – 40 K+ H100/A100 GPUs, one-click deploy • RWA tokenization next → on-chain shares of real-world GPU racks We are scaling the backbone of AI one GPU at a time. #AI #NODEAI #GPU #RWA

Node AI | $GPU

10,220 views • 1 year ago

$15,000 for a Mac sounds insane. Until you realize what Apple is actually selling. 256GB of unified memory. Not a huge GPU farm. Not an NVIDIA rack. Not 4 different machines. Just one box that can keep multiple large LLMs in memory and run them locally. That changes the equation for local AI. NVIDIA wins when you need raw GPU compute. Apple gets interesting when the bottleneck is memory capacity. Because once your model doesn't fit into 24GB, 32GB or 48GB of VRAM, buying a faster GPU doesn't solve the problem. You need more memory. And that's where machines like the Mac Studio start looking less crazy. The interesting question isn't: “Is a $15K Mac faster than an NVIDIA GPU?” It isn't. The real question is: “How much are you willing to pay to make huge local models actually fit?” That’s where unified memory gets interesting.

$15,000 for a Mac sounds insane. Until you realize what Apple is actually selling. 256GB of unified memory. Not a huge GPU farm. Not an NVIDIA rack. Not 4 different machines. Just one box that can keep multiple large LLMs in memory and run them locally. That changes the equation for local AI. NVIDIA wins when you need raw GPU compute. Apple gets interesting when the bottleneck is memory capacity. Because once your model doesn't fit into 24GB, 32GB or 48GB of VRAM, buying a faster GPU doesn't solve the problem. You need more memory. And that's where machines like the Mac Studio start looking less crazy. The interesting question isn't: “Is a $15K Mac faster than an NVIDIA GPU?” It isn't. The real question is: “How much are you willing to pay to make huge local models actually fit?” That’s where unified memory gets interesting.

KyzoroX

19,536 views • 2 months ago