@HPettyIII Profile picture

Harry Petty III

@HPettyIII

Marketing, Product Management, and Business Creation exec specializing in simplifying cloud usage

Similar User
Ravi Balakrishnan photo

@RaviBala8

Frank Palumbo photo

@fpalumbo

Doron Chosnek photo

@dchosnek

Michael Smith photo

@michsmit

Juan Lage photo

@JuanLage

Jason Gmitter photo

@jgmitter

Dan Hanson photo

@danhanso

murali gandluru photo

@muralikrishnag

Craig Huitema photo

@craighuitema

🏎️ @nvidiaaidev developed a new algorithm, TensorRT-LLM MultiShot, which uses NVSwitch to accelerate a key operation, AllReduce, for multi-GPU #inference, making it up to 3X faster than ring-based methods. ⚡Read the tech blog: bit.ly/48wWtBb


xAI's Colossus in Memphis, the world's largest AI supercomputer with 100,000 NVIDIA Hopper GPUs, achieves new heights with NVIDIA Spectrum-X Ethernet. A testament to NVIDIA's dedication to #AI progress. Read more: nvda.ws/4eZbrCd

Tweet Image 1

xAI's Colossus in Memphis, the world's largest AI supercomputer with 100,000 NVIDIA Hopper GPUs, achieves new heights with NVIDIA Spectrum-X Ethernet. A testament to NVIDIA's dedication to #AI progress. Read more: nvda.ws/4eZbrCd

Tweet Image 1

Discover the Llama-3.1-Nemotron-70B-Instruct #LLM, customized by NVIDIA. bit.ly/4dZh6a6

Tweet Image 1

Discover the Llama-3.1-Nemotron-70B-Instruct #LLM, customized by NVIDIA. bit.ly/4dZh6a6

Tweet Image 1

📗 Technical deep dive: Discover how to improve #CFD simulation accuracy while reducing computational costs simulations with #ML and NVIDIA #Modulus. bit.ly/3BKZbqn


NVIDIA's CUDA-Q platform is accelerating #quantum research. A new paper shows how #GPU simulations are enabling research into quantum #machinelearning and drug discovery. Learn more now. bit.ly/4eYhGpE


AlphaFold2-multimer NIM predicts 3D multimeric protein structures in minutes, saving time and money for researchers. No lab required, just accurate and fast computational protein design. #AI #healthcare #biopharma bit.ly/484fwT4

Tweet Image 1

Learn how the right parallelism technique increases #Llama 3.1 405B performance by 1.5x in throughput-sensitive scenarios on an NVIDIA HGX H200 system with NVLink and NVSwitch, and enables a 1.2x speedup in the MLPerf Inference v4.1 Llama 2 70B benchmark. bit.ly/3TYnfw3


NVIDIA’s #GraceCPU Superchip was designed to bring breakthrough performance and efficiency in the data center. Find out what makes #NVIDIAGrace CPU a new kind of #datacenter CPU. Read the latest tech blog. bit.ly/3A6DhNI


Boost your application performance and ROI with our ongoing optimizations of leading #LLMs, designed to deliver high throughput and low latency for real-time demands. Learn more. bit.ly/4eWi3RI


Learn how the right parallelism technique increases #Llama 3.1 405B performance by 1.5x in throughput-sensitive scenarios on an NVIDIA HGX H200 system with NVLink and NVSwitch, and enables a 1.2x speedup in the MLPerf Inference v4.1 Llama 2 70B benchmark. bit.ly/3TYnfw3


#NVIDIABlackwell has arrived. OpenAI, we’re excited to see what you can do with one of the first engineering builds of the #NVIDIADGX B200 system. Learn more about DGX B200 now. bit.ly/404uQNs


Boost your application performance and ROI with our ongoing optimizations of leading #LLMs, designed to deliver high throughput and low latency for real-time demands. Learn more. bit.ly/4eWi3RI


#NVIDIABlackwell has arrived. OpenAI, we’re excited to see what you can do with one of the first engineering builds of the #NVIDIADGX B200 system. Learn more about DGX B200 now. bit.ly/404uQNs


To meet the needs of increasingly large #LLMs, multi-GPU compute is a must for ensuring low-latency and high throughput for users. Learn more about how to leverage NVIDIA NVLink and NVSwitch to achieve the best performance with multi-GPU inference. #AI bit.ly/4dxPzwS


To meet the needs of increasingly large #LLMs, multi-GPU compute is a must for ensuring low-latency and high throughput for users. Learn more about how to leverage NVIDIA NVLink and NVSwitch to achieve the best performance with multi-GPU inference. #AI bit.ly/4dxPzwS


📢 Big news for Linux users! NVIDIA goes fully open-source with GPU kernel modules. Experience enhanced performance and new features like HMM support and confidential computing. 💻 #OpenSource #NVIDIA #LinuxTech bit.ly/3zWl0lP

Tweet Image 1

Read the blog to discover how Livex AI leveraged NVIDIA AI and GKE infrastructure to build #generativeAI agents that provide real-time, human-like customer experiences. ✨➡️ #AI #retail

Tweet Image 1

Loading...

Something went wrong.


Something went wrong.