Results for "memory efficiency"
36 results found
New Technique Losslessly Compresses KV Cache Up to 4x for Faster AI Inference
Speculative KV coding compresses key-value cache up to 4x without loss, potentially cutting memory costs and enabling larger models on existing hardware.
Intel Unveils Massive Memory AI Chip for Data Centers
Intel reveals its next-gen data center GPU with up to 480GB of LPDDR5X memory at Computex.
New AI Architecture Separates Prompts and Reasoning Into Parallel Streams
Researchers propose Multi-Stream LLMs, splitting prompts, thinking and I/O into parallel processes to boost efficiency and reduce latency.
Microsoft Unveils Desktop AI Dev Box That Runs 120B-Parameter Models Locally
Microsoft's Surface RTX Spark Dev Box lets developers run large AI models on local hardware with 128GB unified memory, bypassing cloud costs. The device challenges the per-token pricing model that has dominated AI economics since ChatGPT's launch.
PyTorch Custom Operations Give Developers Deeper Control Over Model Performance
PyTorch's custom operation support lets developers write optimized CUDA kernels, balancing research flexibility with production efficiency.
Apple’s MacBook Neo Faces Fresh Wave of Rivals From Dell and Microsoft
Dell and Microsoft are launching new laptops aimed at the MacBook Neo, but critics say they miss key design and performance lessons from Apple.
Ember.js 7.0 Arrives With Major Rewrite and Modernized Tooling
Ember.js 7.0 introduces a new reactivity system, drops legacy browser support, and improves TypeScript integration. Developers must prepare for breaking changes.
Rust Gains Traction as Python Developers Seek Speed and Safety
Python developers are turning to Rust for performance and safety, with new tools enabling seamless integration.
OpenAI Upgrades ChatGPT Memory for Free Users, Closing Gap With Paid Tiers
OpenAI improved ChatGPT memory, especially for free users. The chatbot now better retains conversation context across sessions. This closes a key gap between free and paid tiers.
Samsung Reveals HBM5 Memory Prototype With In-Package Cooling
Samsung showed its first HBM5 memory prototype at Computex, pairing the next-gen AI memory with a new in-package cooling system called Heat Path Block to tackle thermal challenges.
AI data centers spark memory chip shortage that could raise car and medical device prices
A coalition of nine U.S. trade groups warns the Trump administration that AI-driven demand for DRAM chips is squeezing supply, threatening price hikes across automotive, medical and telecom sectors through 2027.
HP's New Workstation Packs 784GB Memory for Trillion-Parameter AI Models
HP announced the ZGX Fury GB300, a workstation with 784GB unified memory and Nvidia GB300 GPU, handling trillion-parameter AI models. It targets enterprise workloads but at a high price.
Intel Confirms Next-Gen Xeon Server Chips for 2027 With Major Performance Gains
Intel’s Diamond Rapids Xeon CPUs arrive in 2027 with up to 50% more cores and double memory bandwidth to challenge AMD’s EPYC Venice.
Laptop Makers Return to 8GB RAM as Component Costs Bite
Dell and Acer introduced new laptops with 8GB RAM at Computex, reversing the 16GB trend. The shift aims to keep prices low amid ongoing component shortages.
Steam Deck Price Hikes Signal Enduring Hardware Cost Crisis
Valve's Steam Deck price jump of over 40% shows AI demand and geopolitics are pushing hardware costs to a painful new normal for consumers.
AMD Plans Budget-Friendly Re-Release of Ryzen 7 5800X3D CPU
AMD may re-release the Ryzen 7 5800X3D as a 10th Anniversary Edition, offering a powerful upgrade for older AM4 PCs without requiring new DDR5 RAM.
Valorant Anti-Cheat Update Bricks $6,000 Cheating Devices
Riot Games' Vanguard anti-cheat update blocks expensive DMA cheating hardware, turning devices into paperweights. The studio then mocked cheaters on social media.
Rugged Tablet With Built-In Projector Pushes Mobile Boundaries
The 8849 Tank Pad Ultra combines a rugged tablet design with an integrated DLP projector. Its powerful specs and premium price target niche users who need both durability and projection capabilities.
Node.js 26.0.0 Introduces Temporal API for Modern Date Handling
Node.js 26.0.0 is now available, featuring the long-awaited Temporal API. This update modernizes date and time management for developers.
RAM Prices Stay High, but PC Upgrades Shift to Peripherals
With RAM costs still elevated, PC enthusiasts are turning to upgrades like keyboards, monitors and network gear. Memorial Day sales highlight the trend.