NVIDIA

Quote Tweet: https://twitter.com/i/status/1963255671660441911

AI Blueprints + #NVIDIARTXPRO = Lightning-fast 3D scene building 🎨

👀 See what the hype is about: http://nvda.ws/463daTM

⚡ Learn how #NVIDIARTXPRO 6000 Blackwell is accelerating compute across:

🔬 Advanced simulation
🌌 Scientific computing
🧬 Genomics

Learn more: https://nvda.ws/4ncfZJp

Lights, camera, AI! 🎬🤖

Watch NVIDIA + @DellTech + @qvestgroup dive into how AI is transforming broadcast workflows—think real-time content search, auto video summarization, and global content localization. 🌍🎙️

Catch it here: https://nvda.ws/4oXOd4Q

# Alt Text: A sleek black NVIDIA RTX 4000 SFF (Small Form Factor) GPU card displayed against a dark background with text reading "THE SUMMER I LAUNCHED SFF GPUs." The compact half-size graphics card features a prominent circular cooling fan and is positioned at an angle, showcasing its slim design and professional engineering.

☀️ The summer we launched SFF GPUs...

⚡ We packed Blackwell power into half-size cards
🤖 We ran AI, CAD & 3D workflows way faster
🔋 We did it all at just 70W max

Catch the full story: https://nvda.ws/3HjTMd6

Curious what the most powerful desktop GPU ever created looks like? 👀

🔍 Take a closer look: https://nvda.ws/47O6Py2

Quote Tweet: https://twitter.com/i/status/1954931074942837100

ICYMI: Our new GPUs fit in your pocket 🤏(kinda)

Meet the NVIDIA RTX PRO 4000 SFF & 2000—compact design, mighty performance.

🔗 https://nvda.ws/3HjTMd6

AI factories just got an upgrade.⚡

In the latest episode of the Reshaping Workflows podcast, Jeremy Williford dives into how NVIDIA + @DellTech built Dell AI Factory 2.0 ✨

Spoiler: it’s a game-changer: https://nvda.ws/45VA3d5

Your footage called. It wants more speed. 😉

See how NVIDIA RTX PRO Blackwell GPUs handle 4:2:2 10-bit video, AI magic, and lightning-fast encoding without breaking a sweat 🎥

🔗 https://nvda.ws/45BnQJk

# Alt Text NVIDIA RTX PRO 4000 Blackwell SFF Edition graphics card displayed against a black background with flowing wave patterns. The compact, rectangular card features a circular cooling fan on top, vertical heat sink fins along the side, and the NVIDIA logo. Text "RTX PRO" appears in the bottom right corner.

The second star of our #SIGGRAPH2025 lineup: NVIDIA RTX PRO 4000 Blackwell SFF Edition 🌟

⚡24GB of GDDR7 memory
⚡Space-saving design
⚡Packed with power for AI, rendering, and more.

🔗 Check it out: https://nvda.ws/41Hi1bX

Explore the 22-qubit demo and see #CUDAQ in action: https://nvidia.github.io/cuda-quantum/latest/applications/python/skqd.html?ncid=so-twit-844277

# Alt Text Bar chart comparing average execution time for postprocessing with k=11. GPU completes in 0.1373 seconds while CPU takes 19.4353 seconds, with a callout indicating GPU is 141.6x faster.

NVIDIA GPUs are accelerating #QuantumComputing workflows. ⚡

Researchers working on Sample Based Krylov Quantum Diagonalization (SBKQD) methods can now use our NVIDIA CUDA-Q platform to perform ground-state energy estimation without the issue of ‘Barren Plateaus’.

🔗 Dive into the benchmarks and code samples here: https://developer.nvidia.com/blog/advanced-large-scale-quantum-simulation-techniques-in-cuquantum-sdk-v25-11/?ncid=so-twit-654891&linkId=100000398133283

Quantum developers can now achieve up to 100x speedups for validating large scale utility circuits, and 1000x speedups for simulating #QuantumErrorCorrection codes. ✨

Advanced Large-Scale Quantum Simulation Techniques in cuQuantum SDK v25.11 | NVIDIA Technical Blog Simulating large-scale quantum computers has become more difficult as the quality of quantum processing units (QPUs) improves. Validating the results is key to ensure that after the devices scale…

The new #cuQuantum SDK v25.11 is here, introducing cuPauliProp and cuStabilizer to supercharge large-scale #QuantumComputing simulations on NVIDIA GPUs. ⚡️

Professor Hao Zhang says DGX B200 “enables us to prototype and experiment much faster than using previous-generation hardware,” with performance among the best in the world.

We look forward to seeing the results of their research and projects. 🙌

UC San Diego Lab Advances Generative AI Research With NVIDIA DGX B200 System The Hao AI Lab research team at the University of California San Diego recently received an NVIDIA DGX B200 system to elevate their work in LLM inference.

🌌 @UCSanDiego's Hao AI Lab now has full access to NVIDIA DGX B200 at the San Diego Supercomputer Center, opening up new research opportunities across campus ➡️ https://blogs.nvidia.com/blog/ucsd-generative-ai-research-dgx-b200/

Read the full technical deep dive: https://developer.nvidia.com/blog/real-time-decoding-algorithmic-gpu-decoders-and-ai-inference-enhancements-in-nvidia-cuda-q-qec/?ncid=so-twit-566140&linkId=100000397936187

#QPU builders and researchers can now run AI and GPU-accelerated decoders for both real and simulated QPUs, scaling #quantum hardware into useful quantum-GPU supercomputers.

# Alt Text A glowing green atomic nucleus at the center surrounded by white orbital electron paths against a dark gray background, symbolizing quantum computing and atomic physics. The atom features geometric coordinate axes overlaid on the nucleus, representing quantum error correction and computational modeling.

Real-time decoding is the key challenge facing #quantumerrorcorrection, and now it can be performed with the latest release of NVIDIA CUDA-Q QEC. 🛠️⚛️

Learn how to leverage this framework in our latest tech blog. 👉 https://developer.nvidia.com/blog/using-ai-physics-for-technology-computer-aided-design-simulations/?ncid=so-twit-105750

# Alt Text: Close-up of semiconductor wafer processing equipment showing a large circular wafer with a grid pattern of microchip structures mounted on an automated handling apparatus with white mechanical components and metal framework.

✨ @SKhynix is pioneering AI Physics using NVIDIA #PhysicsNeMo to create high-fidelity surrogate models, accelerating semiconductor design from hours to milliseconds. 🤯

https://www.nature.com/articles/s41467-025-65836-3?linkId=100000396517062&error=cookies_not_supported&code=87a19b0e-2c20-4f2b-a697-ea94f79341ab

Dive into our Nature Communications review to see how state-of-the-art AI is accelerating everything from qubit design and control to real-time error correction, paving the way for useful fault-tolerant QC.

Artificial intelligence for quantum computing - Nature Communications Quantum computing devices of increasing complexity are becoming more and more reliant on automatised tools for design, optimization and operation. In this Review, the authors discuss recent developments in “AI for quantum", from hardware design and control, to circuit compiling, quantum error correction and postprocessing, and discuss future potential of quantum accelerated supercomputing, where AI, HPC, and quantum technologies converge.

The path to scalable #quantumcomputing is being supercharged by AI. 💡

🌌 We provide detailed visibility into application algorithms, so you can fine tune performance where it matters most, with NVIDIA Nsight Systems, a system-wide performance analysis tool.

➡️ https://developer.nvidia.com/nsight-systems?ncid=so-twit-907069

The @IBM Storage Scale System 6000 and NVIDIA Spectrum-X networking deliver top performance across storage and networking layers, helping enterprises accelerate #HPC, AI, and data intensive workloads.

⚡ High-performance storage and deep application insight are the foundation of true optimization.

CUDA Programming Guide: Everything CUDA for novices and experts.

🔗 Download the toolkit and see the next era of GPU programming: https://developer.nvidia.com/blog/nvidia-cuda-13-1-powers-next-gen-gpu-programming-with-nvidia-cuda-tile-and-performance-gains?ncid=so-twit-632512&linkId=100000395867159

✅ Announcing CUDA Tile: Future-proof kernels by abstracting Tensor Cores.
✅ Performance Wins: Up to 4x speed-up with cuBLAS Grouped GEMM & 2x in cuSOLVER on Blackwell.
✅ Green Contexts: Finer GPU resource control via Runtime API.
✅ New Tooling: Nsight Compute now profiles CUDA Tile kernels.
✅ New

# Alt Text: 3D visualization of a glowing neural network or tensor structure inside a wireframe cube. The center features a bright spiral pattern in lime green and cyan, surrounded by flowing data represented in gradient colors from green to purple, against a black background. The image represents CUDA Tile technology for GPU computing and accelerated processing.

🚨 NVIDIA CUDA 13.1 brings the largest update to the #CUDA platform in 20 years.

Discover the new features and updates for improving performance and driving #acceleratedcomputing, including: