Cerebras Runs GPT-5.6 Sol at 750 Tokens/Second: Real-Time Frontier AI Finally Arrives
Cerebras and OpenAI launch Ultrafast tier, delivering GPT-5.6 Sol at 750 tokens/second — 14× faster than standard. Real-time frontier intelligence is here.
Hardware
OpenAI Unveils Jalapeño: The First Custom AI Chip That Shifts the Hardware Power Balance
OpenAI’s first custom chip, Jalapeño, marks a turning point in the AI hardware wars—reducing reliance on Nvidia and signaling a fully integrated future for the company.
Etched's Sohu Chip: A Transformer-Specific Bet That Just Raised $800 Million
Etched emerged from stealth today with $800 million in funding and over $1 billion in signed customer contracts, unveiling its Sohu chip – an ASIC designed exclusively for transformer inference. The move signals a growing push to build domain-specific hardware for large language models, potentially reshaping the AI chip market.
IBM's Sub-1nm Chip Breaks Silicon Limits, Promises 50% Performance Leap
IBM announces a 0.7nm chip with 100 billion transistors, doubling density over its 2nm node and delivering up to 50% more performance or 70% lower power.