PyTorch
10 analyses mentioning PyTorch, newest first.
Google & Meta Open-Source TorchTPU, Bringing Native PyTorch Support to TPUs
Google and Meta launched TorchTPU, a native PyTorch backend for Google's TPUs, aiming to simplify model portability across different accelerators like GPUs and CPUs.
Hugging Face Launches Tool to Test Model Compatibility on AWS Trainium
Hugging Face introduced a Space allowing users to check if popular Transformer models can run unmodified on AWS Trainium hardware, providing necessary scripts and compatibility details.
Google and Amazon Challenge Nvidia Dominance with Open PyTorch Forks for Custom AI Chips
Google released TorchTPU, an open-source PyTorch version for its TPUs, joining Amazon's effort to break Nvidia's hardware monopoly in AI development.
Hugging Face Unlocks Native TPU Support for PyTorch, Simplifying AI Deployment
Hugging Face's TorchTPU brings native PyTorch backend support for Google's TPUs, allowing seamless training and inference for models like Transformers and Diffusers.
PyTorch 2.15 Supercharges Mac AI: FP8, MoE Speedups, and Metal Native Kernels
PyTorch 2.15 significantly boosts Mac performance for AI by bringing FP8 support, drastically speeding up MoE decoding, and migrating more operations to native Metal kernels.
Amazon Launches TorchNeuron to Challenge Nvidia's AI Software Lock-In
AWS released TorchNeuron, a tool enabling easy deployment of PyTorch-based AI apps onto Amazon's Trainium chips, directly challenging Nvidia's dominance in the AI software stack.
TorchNeuron Revolutionizes HF Model Deployment on AWS Trainium with Native PyTorch Support
TorchNeuron brings native PyTorch support, including eager execution and FSDP, to AWS Trainium, simplifying the deployment of Hugging Face models and making the platform more accessible for advanced AI workflows.
New Loss Function Makes Distilling LLMs Viable for Ultra-Long Context Training
A novel, memory-efficient loss function allows practitioners to train smaller student models on massive context windows using advanced knowledge distillation techniques.
Demystifying PyTorch Profiling: A Deep Dive into CUDA Overhead and Kernel Optimization
This beginner's guide introduces torch.profiler, explaining how to interpret complex execution traces to move deep learning models from overhead-bound to compute-bound regimes.
Thinking Machines Lab Aggregates Elite Talent and Multi-Billion Dollar Cloud Deals, Positioning as Major AI Player
TML is rapidly establishing itself as a serious competitor to established players like Meta and Anthropic, fueled by high-profile talent acquisition and major cloud infrastructure deals.