Tech Times

DeepSeek V4 Architecture: How Sparse Attention Cuts Inference Costs, What NIST Found

DeepSeek V4 architecture uses sparse attention to cut inference costs 73% at one-million-token contexts, but a NIST ...
CNBC

Microsoft unveils new AI models to lessen reliance on OpenAI and lower costs for developers

At its Build developer conference in San Francisco, Microsoft announced MAI-Code-1-Flash, its inaugural model in the AI coding space. Microsoft is trying to establish a presence with proprietary ...
The New York Times

A Cattle Ranch Is Doing What the Ivy League Can’t

Ms. Leibowitz is an editor at Times Opinion. For this essay, she spoke to two dozen students, teachers and alumni and visited Deep Springs College. May 20, 2026 When I was in college in the 2010s, ...
IEEE

A Comparative Study of Deep Learning models for Garbage Classification

Abstract: The rising amount of waste generated worldwide makes it an imperative to apply intelligent systems and efficient waste classification to enable recycling. Manual waste classification is time ...
Reuters

DeepSeek's new AI model does not wow markets in fast-changing industry

DeepSeek-V4 shows improvement but faces strong competition from Kimi and Qwen, data shows Reaction muted as rapid AI advances, new entrants expected China's push for AI self-reliance has taken market ...
TechCrunch

DeepSeek previews new AI model that ‘closes the gap’ with frontier models

Chinese AI lab DeepSeek has launched two preview versions of its newest large language model, DeepSeek V4, a much-awaited update to last year’s V3.2 model and the accompanying R1 reasoning model that ...
MIT Technology Review

Three reasons why DeepSeek’s new model matters

The long-awaited V4 is more efficient and a win for Chinese chipmakers. On April 24, Chinese AI firm DeepSeek released a preview of V4, its long-awaited new flagship model. The model can process much ...
VentureBeat

OpenAI launches Privacy Filter, an open source, on-device data sanitization model that removes personal information from enterprise datasets

Credit: VentureBeat made with OpenAI ChatGPT-Images-2.0 In a significant shift toward local-first privacy infrastructure, OpenAI has released Privacy Filter, a ...
TechCrunch

Meta debuts the Muse Spark model in a ‘ground-up overhaul’ of its AI

Meta released an AI model on Wednesday called Muse Spark, which marks its “first step” toward an “overhaul of [its] AI efforts.” Muse Spark is the inaugural model to come out of Meta Superintelligence ...
EurekAlert!

Hybrid modeling strategy based on deep learning surrogate models enables accurate multi-objective optimization of iso-octanol oxidation

The selective oxidation of alcohols to aldehydes is a vital reaction in organic synthesis and chemical production. However, this reaction exhibits a complex reaction network influenced by temperature, ...
Mashable

Google launches Gemma 4, a new open-source model: How to try it

Google just released its newest AI model Gemma 4, which is now both open and open source. Credit: Thomas Fuller/SOPA Images/LightRocket via Getty Images Google just released the latest version of its ...
Ars Technica

Running local models on Macs gets faster with Ollama’s MLX support

Ollama, a runtime system for operating large language models on a local computer, has introduced support for Apple’s open source MLX framework for machine learning. Additionally, Ollama says it has ...