Hardware

OpenAI Uses Its Own LLMs to Design Jalapeño Chip

IEEE Spectrum AI • 2 days ago

NVIDIA Groq 3 LPX Boosts Vera Rubin Power Efficiency

NVIDIA Developer Blog • 1 day ago

NVIDIA Boosts DeepSeek-V3 Training Speed by 10x

NVIDIA Developer Blog • 2 days ago

Meta Brings Low-Precision FlashAttention-4 to Blackwell

PyTorch Blog • 18 hrs ago

Edge0 Runs 35B Qwen AI Model on iPhone with 2.9 GiB RAM

AlphaSignal • 4 days ago

Perplexity Cuts Search Latency by 82% With CobbleDB

AlphaSignal • 1 day ago

Nvidia Jetson AGX Thor Speeds Up Qwen3.6 by 6.4x

NVIDIA Developer Blog • 16 hrs ago

Agility Robotics Unveils Safe Digit 5 Humanoid

Ars Technica AI • 1 day ago

Agility Robotics Unveils Digit 5 Humanoid Robot

AI Business • 16 hrs ago

Apple Plans AI Server Return Using Nvidia Tech

The Verge AI • 19 hrs ago

Meta and OpenAI Drive Massive Data Center Power Demand

WIRED AI • 4 days ago

Snap Launches Specs Intelligence AI Assistant on iOS

The Verge AI • 17 hrs ago

Unitree leads cheap humanoid robot market via cost-cutting

Ars Technica AI • 2 days ago

Meta Designs Luna Smart Glasses Without Built-In Cameras

TechCrunch AI • 17 hrs ago

NVIDIA NVLink 6 Prevents Packet Loss in AI Factories

NVIDIA Developer Blog • 1 day ago

Bojie Li Releases Open-Source AI Infrastructure Book

AlphaSignal • 1 day ago

Fluidic Logic Systems Offer Silicon-Free Computing

The Chip Letter • 4 days ago

IBM Debuts Nighthawk r2 to Advance Quantum Error Correction

IBM Research AI • 1 day ago

Regent Craft flies its electric Seaglider with crew

Superhuman AI • 3 days ago

OpenAI Uses Its Own LLMs to Design Jalapeño Chip

OpenAI has designed its debut Jalapeño AI chip in under 20 months using its own LLMs, proving that generative AI can radically accelerate the...

IEEE Spectrum AI|2 days ago

NVIDIA Groq 3 LPX Boosts Vera Rubin Power Efficiency

NVIDIA is integrating the Groq 3 LPX accelerator into its Vera Rubin NVL72 platform to dramatically boost power efficiency and throughput for massive...

NVIDIA Developer Blog|1 day ago

NVIDIA Boosts DeepSeek-V3 Training Speed by 10x

NVIDIA integrated its Transformer Engine with JAX to achieve a 10.4x throughput boost training DeepSeek-V3, overcoming a major communication...

NVIDIA Developer Blog|2 days ago

Meta Brings Low-Precision FlashAttention-4 to Blackwell

Meta researchers have extended FlashAttention-4 with end-to-end MXFP8 support for Nvidia Blackwell GPUs, unlocking massive speedups for large...

PyTorch Blog|18 hrs ago

Edge0 Runs 35B Qwen AI Model on iPhone with 2.9 GiB RAM

Edge0 has launched an open-source framework that runs a 35-billion-parameter AI model on an iPhone with just 2.9 GiB of active memory, bringing...

AlphaSignal|4 days ago

Perplexity Cuts Search Latency by 82% With CobbleDB

Perplexity has introduced CobbleDB, a custom Rust-based key-value store that reduces search read latency by 82% and lowers infrastructure costs by an...

AlphaSignal|1 day ago

Nvidia Jetson AGX Thor Speeds Up Qwen3.6 by 6.4x

Nvidia's TensorRT Edge-LLM ran the Qwen3.6-27B model 6.4 times faster than a standard reference on Jetson AGX Thor, showing that complex AI agents...

NVIDIA Developer Blog|16 hrs ago

Agility Robotics Unveils Safe Digit 5 Humanoid

Agility Robotics has introduced Digit 5, a humanoid robot designed to safely work alongside humans in warehouses and factories without the need for...

Ars Technica AI|1 day ago

Agility Robotics Unveils Digit 5 Humanoid Robot

Agility Robotics has unveiled Digit 5, a humanoid robot designed to work safely alongside humans in warehouses, aiming to overcome the safety...

AI Business|16 hrs ago

Apple Plans AI Server Return Using Nvidia Tech

Apple reportedly plans to re-enter the server market by 2029, leveraging its future M8 Ultra chips and Nvidia technology to capture a share of the...

The Verge AI|19 hrs ago

Meta and OpenAI Drive Massive Data Center Power Demand

As tech giants like Meta and OpenAI transition from simple chatbots to autonomous AI agents, the massive energy demands of these persistent systems...

WIRED AI|4 days ago

Snap Launches Specs Intelligence AI Assistant on iOS

Snap has introduced Specs Intelligence, an anticipatory AI assistant designed to connect user accounts and proactively manage daily tasks across iOS,...

The Verge AI|17 hrs ago

Unitree leads cheap humanoid robot market via cost-cutting

Chinese robotics firm Unitree has captured the low-cost humanoid market with its G1 and R1 models, but intense founder micromanagement and a US...

Ars Technica AI|2 days ago

Meta Designs Luna Smart Glasses Without Built-In Cameras

Meta is reportedly developing a new pair of camera-free smart glasses codenamed Luna, aiming to bypass growing privacy concerns while keeping users...

TechCrunch AI|17 hrs ago

NVIDIA NVLink 6 Prevents Packet Loss in AI Factories

NVIDIA has detailed the multi-layer resiliency architecture of its NVLink 6 interconnect, which eliminates packet loss and slashes recovery times to...

NVIDIA Developer Blog|1 day ago

Bojie Li Releases Open-Source AI Infrastructure Book

Bojie Li has released an open-source textbook on AI infrastructure to help engineers optimize large language model training and inference by...

AlphaSignal|1 day ago

Fluidic Logic Systems Offer Silicon-Free Computing

While silicon dominates modern processing, fluidic logic systems that compute using air and water offer a highly durable alternative for robotics and...

The Chip Letter|4 days ago

IBM Debuts Nighthawk r2 to Advance Quantum Error Correction

IBM has released its Quantum Nighthawk r2 platform alongside new Qiskit tools to help researchers bridge the gap between quantum error mitigation and...

IBM Research AI|1 day ago

Regent Craft flies its electric Seaglider with crew

Regent Craft has completed the first crewed flight of its electric Seaglider, marking a major milestone for high-speed, zero-emission coastal transit.

Superhuman AI|3 days ago