OpenAI sidesteps Nvidia with unusually fast coding model on plate-sized chips

OpenAI has spent the past year systematically reducing its dependence on Nvidia. The company signed a massive multi-year deal with AMD in October 2025, struck a $38 billion cloud computing agreement ...
Interesting Engineering

GPT-5.3-Codex-Spark delivers ultra-fast real-time AI coding at 1,000 tokens per second

OpenAI has launched GPT-5.3-Codex-Spark, its first AI model built specifically for real-time coding, capable of generating more than 1,000 tokens per second while handling real-world software ...

OpenAI's new Spark model codes 15x faster than GPT-5.3-Codex - but there's a catch

OpenAI's new GPT-5.3-Codex-Spark promises ultra-fast, conversational AI coding, if you can tolerate a few trade-offs.
Every on MSN

How Claude Code is transforming finance—without turning you into a coder

by Brooker Belcourt TL;DR: As the head of financial services consulting at Every, Brooker Belcourt has helped take top hedge funds, asset managers, and research teams from AI-curious to AI-native.
IEEE

A Low-Power 8-bit Spintronic Rotary Encoder With Single-Track Gray Code Based on Spin-Hall Effect-Driven Domain Wall Motion

Abstract: This article introduces a low-power spintronic rotary encoder capable of 8-bit analog-to-digital conversion. The proposed encoder leverages spin-Hall effect-driven domain wall motion ...
Communications of the ACM

Formal Reasoning Meets LLMs: Toward AI for Mathematics and Verification

A marriage of formal methods and LLMs seeks to harness the strengths of both.
IEEE

On the I/O Cost of Linear Repair Schemes for Arbitrary $(n,k)$ Reed-Solomon Codes

Abstract: The I/O cost, i.e., the total number of symbols to be read during the single node failure/repair process in a distributed storage system, is one of the most important metrics in repairing ...