716 place 0

802 IndexCache, a new sparse attention optimizer, delivers 1.82x faster inference on long-context AI models

VentureBeat
VentureBeat 1 place · 03/26/2026 20:00 EDT

Processing 200,000 tokens through a large language model is expensive and slow: the longer the context, the faster the costs spiral. Researchers at Tsinghua University and Z.ai have built a technique called IndexCache that cuts up to 75% of the redundant computation in sparse attention models, delivering up to 1.82x faster time-to-first-token and 1.48x faster generation throughput at that context length.The technique applies to models using the DeepSeek Sparse Attention architecture, including the latest De

To see detailed statistics for the news please log in »

Read the original

Add your comment
You must be logged in with Facebook to read and write comments.

A newsletter a day!

You may get 10 most important news around midday in daily newsletter. Press the button and we will send you the most important news only, no spam attached.

or register

LIKE us on Facebook so you won't miss the most important news of the day!

News from the same source
VentureBeat VentureBeat
Silicon Valley
George Avalos @ Silicon Valley 1 place · 02/07/2106 01:28 EDT

Newark apartment complex bought for much less than prior value

An East Bay apartment complex has been bought at a price that's well below its prior value. Read more

0

🔮
27.03.2026 ♏︎ Dear Scorpio! Today may bring you many challenges and minor disappointments related to various areas... Read more ›
Silicon Valley
George Avalos @ Silicon Valley 2 place · 02/07/2106 01:28 EDT

PG&E buys San Jose building to bolster South Bay operations

A PG&E Corp. unit has bought a San Jose building in a move to bolster the utility's South Bay operations. Read more

0

Digital Trends
Manisha Priyadarshini @ Digital Trends 1 place · today 17:30 EDT

Apple says Lockdown Mode thwarted spyware attacks with a clean slate

Apple says it has not seen a single successful spyware attack on iPhones with Lockdown Mode enabled, highlighting how the feature limits attack surfaces and blocks common entry points used by advanced threats. Read more

0 newcommer

Habr
EkaterinaRabynina (INFOSTART.RU) @ Habr 1 place · today 17:30 EDT

Какими инструментами пользуется бизнес-аналитик в 2026 году

Современный бизнес-аналитик 1С занимается не только сбором требований заказчика и передачей их разработчику. Эта роль стала шире: здесь требуются и навыки проектного управления, и понимание архитектуры решений.Наталья Китавина, аналитик проектов 1С Ресурсного центра Инфостарта, рассказывает, какими программами сегодня пользуются аналитики и как подобрать инструментальный стек под свои рабочие задачи. Читать далее Read more

0 newcommer

Engadget
Ian Carlos Campbell @ Engadget 1 place · today 17:26 EDT

Kash Patel's personal email account was accessed by hackers linked to Iran

A hacking group called Handala has gained access to FBI Director Kash Patel's email account, Reuters reports. The group published content from Patel's email on their website as proof, including photos of Patel "sniffing and smoking cigars" and "making a face while taking a picture of himself in the mirror with a ​large bottle of rum." TechCrunch was able to independently confirm that at least some of the emails Handala... Read more

0 newcommer

BetaKit
Jesse Cole @ BetaKit 1 place · today 17:20 EDT

More than $6 million in federal defence funding bound for Alberta

Funding will support Edmonton’s Wyvern and the establishment of a new commercialization centre at the U of A. Read more

0 fresh

Engadget
Karissa Bell @ Engadget 2 place · today 17:17 EDT

Mark Zuckerberg offered to 'help' Elon Musk with DOGE in 2025

Elon Musk and Mark Zuckerberg have a complicated history. In 2023, the two vowed to fight each other in a cage match that never happened. But by early 2025, when both were cozying up to the newly-elected President Donald Trump, they were apparently on more friendly terms. In February of that year, Zuckerberg texted Musk approvingly about his work with the now-defunct Department of Government Efficiency (DOGE). "Looks like DOGE... Read more

0 fresh

SlashGear
SlashGear 1 place · today 17:15 EDT

Can An HDMI Cable Go Bad? Here's What You Need To Know

If you're having trouble with your TV or device not displaying images properly over HDMI, don't forget to check if the HDMI cable itself has died. Here's why. Read more

0 fresh

Gizmodo
Ece Yildirim @ Gizmodo 1 place · today 17:10 EDT

Epstein Victims Sue Google, Claim AI Mode Exposed Personal Information

Google's AI republished sensitive info like contact information, the suit claims. Read more

0 fresh

GSMArena.com
GSMArena.com 1 place · today 17:09 EDT

OnePlus Nord CE6 leaks in hands-on image, Nord CE6 Lite is on the way too

OnePlus is launching the Nord 6 on April 7. While this device has been leaking a lot lately, until today we weren't sure whether there will be a Nord CE6 too. According to a new leak from a tipster over on X, the answer is, thankfully, yes. Not just that, but the image below purportedly shows the upcoming Nord CE6 for the first time ever. The design is clearly reminiscent... Read more

0 fresh

Silicon Canals
Christian Kelly @ Silicon Canals 1 place · today 17:07 EDT

Age bans won’t save kids from social media. Design mandates might

A landmark US jury verdict finding Meta and Google negligent in harming minors has intensified a fragmented global response — revealing deep divides in how nations assign blame, enforce compliance, and grapple with the uncomfortable psychology of screen-dependent parenting. Read more

0 fresh

Gizmodo
Isaiah Colbert @ Gizmodo 2 place · today 17:00 EDT

‘Wicked Spot’ Is a Fun, Sapphic Rom-Com That Yeets a Witch Into the Magical World of Influencer Culture

If you like 'Green Yuri,' check out Sal Jiang's new manga about a witch's foray into the magic of social media likes, thirst traps, and an enemies-to-lovers relationship with her internet troll. Read more

0 fresh

Slashdot
BeauHD @ Slashdot 1 place · today 17:00 EDT

Windows PCs Crash Three Times As Often As Macs, Report Says

A workplace-device study says Windows PCs crash significantly more often than Macs, lag further behind on patching and encryption in some sectors, and are typically replaced sooner. TechSpot reports: Omnissa's 2026 State of Digital Workspace report outlines the IT challenges that various organizations face from the growing use of AI and the heterogeneous deployment of enterprise devices. The relative instability of Windows and Android is a recurring theme throughout the... Read more

0 fresh

MacRumors
Joe Rossignol @ MacRumors 1 place · today 16:53 EDT

Apple to Launch These 15+ New Products Later This Year

March has been an incredibly busy month for Apple, with the company unveiling more than 10 new products and accessories. We said hello to the MacBook Neo at the start of the month, and we bid farewell to the Mac Pro at the end of it. Beyond the usual annual updates to iPhones and Apple Watches, Apple's all-new smart home hub is finally expected to launch later this year, once... Read more

0 fresh

CNET
Jesse Orrall @ CNET 1 place · today 16:47 EDT

4 Different Futures for Quantum Computing Converge at Nvidia GTC

Four different quantum systems with four different types of qubits were on display at Nvidia GTC, here's how they all work. Read more

0 fresh

The most popular news from the same source for the last week
VentureBeat VentureBeat
VentureBeat
VentureBeat 3 place · 03/22/2026 12:00 EDT

Look, we've spent the last 18 months building production AI systems, and we'll tell you what keeps us up at night — and it's not whether the model can answer questions. That's table stakes now. What haunts us is the mental image of an agent autonomously approving a six-figure vendor contract at 2 a.m. because someone typo'd a config file.We've moved past the era of "ChatGPT wrappers" (thank God), but... Read more

0

VentureBeat
VentureBeat 3 place · 03/22/2026 15:00 EDT

Not long ago, the idea of being a “generalist” in the workplace had a mixed reputation. The stereotype was the “jack of all trades” who could dabble in many disciplines but was a “master of none.” And for years, that was more or less true. Most people simply didn’t have access to the expertise required to do highly cross-functional work. If you needed a new graphic, you waited for a... Read more

0

VentureBeat
VentureBeat · 03/23/2026 00:00 EDT

The $29.3 billion AI coding tool just got caught with its provenance showing. When Cursor launched Composer 2 last week — calling it "frontier-level coding intelligence" — it presented the model as evidence that the company is a serious AI research lab, not just a forked integrated development environment (IDE) wrapping someone else's foundation model. What the announcement omitted was that Composer 2 was built on top of Kimi K2.5,... Read more

0

VentureBeat
VentureBeat · 03/23/2026 00:00 EDT

Presented by Tulsa Innovation LabsAs the global energy system evolves, companies are racing to adopt technologies that can deliver real-world solutions, especially in hard-to-abate industries. Oklahoma, long known as the oil capital of the world, is a center for energy innovation, with Rose Rock Bridge at the forefront.A non-profit based in Tulsa, Rose Rock Bridge is a pilot deployment studio that connects early-stage energy startups with corporate energy partners, non-dilutive... Read more

0

VentureBeat
VentureBeat · 03/23/2026 04:00 EDT

The AI image generation market has had an uncontested leader for months. Google's Nano Banana family of models has set the standard for quality, speed, and commercial adoption, while competitors from OpenAI to Midjourney have jockeyed for second place. That hierarchy shifted on Sunday when Luma AI, a startup better known for its Dream Machine video generation tool, publicly released Uni-1 — a model that doesn't just compete with Google... Read more

0

VentureBeat
VentureBeat · 03/23/2026 07:30 EDT

The prevailing assumption in AI development has been straightforward: larger models trained on more data produce better results. Nvidia's latest release directly challenges that size assumption — and the training recipe behind it may matter more to enterprise AI teams than the model itself. The open-weight model's Cascade RL post-training pipeline, detailed in Nvidia's technical report, offers a reproducible blueprint for enterprise teams building domain-specific reasoning systems without training from... Read more

0

VentureBeat
VentureBeat · 03/23/2026 12:00 EDT

Look, we've spent the last 18 months building production AI systems, and we'll tell you what keeps us up at night — and it's not whether the model can answer questions. That's table stakes now. What haunts us is the mental image of an agent autonomously approving a six-figure vendor contract at 2 a.m. because someone typo'd a config file.We've moved past the era of "ChatGPT wrappers" (thank God), but... Read more

0

VentureBeat
VentureBeat · 03/23/2026 12:00 EDT

Getting AI agents to perform reliably in production — not just in demos — is turning out to be harder than enterprises anticipated. Fragmented data, unclear workflows, and runaway escalation rates are slowing deployments across industries.“The technology itself often works well in demonstrations,” said Sanchit Vir Gogia, chief analyst with Greyhound Research. “The challenge begins when it is asked to operate inside the complexity of a real organization.” Burley Kawasaki,... Read more

0

VentureBeat
VentureBeat · 03/23/2026 15:00 EDT

Not long ago, the idea of being a “generalist” in the workplace had a mixed reputation. The stereotype was the “jack of all trades” who could dabble in many disciplines but was a “master of none.” And for years, that was more or less true. Most people simply didn’t have access to the expertise required to do highly cross-functional work. If you needed a new graphic, you waited for a... Read more

0

VentureBeat
VentureBeat · 03/23/2026 17:30 EDT

The Innovation Showcase is back at Transform 2026: The Orchestration of Enterprise Agentic AI at Scale, taking place July 14 and 15 in Menlo Park.This year, we are moving beyond generative AI to autonomous agents, focusing on enterprise agentic orchestration, LLM observability and evaluation (LLMOps), RAG infrastructure, inference platforms and optimization, and agentic AI security and identity.We’re on the hunt for the 10 most innovative autonomous agent technologies poised to... Read more

0

Most popular sources

  • You see 865 news out of 865.
  • Sources 61 out of 61.
VentureBeat 0%
Startup News 0%
Tech Wire Asia 0%
ArcticStartup 0%
Ubergizmo 0%
View sources »

LIKE us on Facebook so you won't miss the most important news of the day!

27.03.2026 17:49
Last update: 17:35 EDT.
News rating updated: 23:41.

What is Times42?

Times42 brings you the most popular news from tech news portals in real-time chart.
Read about us in FAQ section.


Times42 © 2026