Benchmarks comparing inference performance of the NVIDIA GH200 Grace Hopper Superchip, enhanced by ZeRO-Inference, to NVIDIA H100 and A100 Tensor Core GPUs.
Benchmarks on NVIDIA’s Transformer Engine, which boosts FP8 performance by an impressive 60% on GPT3-style model testing on NVIDIA H100 Tensor Core GPUs.
GPU benchmarks on Lambda’s offering of the NVIDIA H100 SXM5 vs the NVIDIA A100 SXM4 using DeepChat’s 3-step training example.
How to use FlashAttention-2 on Lambda Cloud, including H100 vs A100 benchmark results for training GPT-3-style models using the new model.
Learn how to use mpirun to launch a LLaMA inference job across multiple cloud instances if you do not have a multi-GPU workstation or server.
Lambda and Hugging Face are collaborating on a 2-week sprint to fine-tune OpenAI's Whisper model in as many languages as possible.
RTX 4090 vs RTX 3090 benchmarks to assess deep learning training performance, including training throughput/$, throughput/watt, and multi-GPU scaling.
This article discusses the performance and scalability of H100 GPUs and the whys for upgrading your ML infrastructure with the H100 release from NVIDIA.
This tutorial summarizes how to write and launch PyTorch distributed data parallel jobs across multiple nodes, with working examples with the torch.distributed.launch, torchrun and mpirun APIs.
This blog describes how to set up a Run:AI cluster on Lambda Cloud with one or multiple cloud instances.
While waiting for NVIDIA's next-generation consumer & professional GPUs, here are the best GPUs for Deep Learning currently available as of March 2022.
GPU benchmarks on NVIDIA A40 GPUs with 48 GB of GDDR6 VRAM, including performance comparisons to the NVIDIA V100, RTX 8000, RTX 6000, and RTX 5000.
This post discusses the Total Cost of Ownership (TCO) for a variety of Lambda A100 servers and clusters. We calculate the TCO for individual Hyperplane-A100 servers, compare the cost with renting a AWS p4d.24xlarge instance, and walk through the cost of building and operating A100 clusters.
...