Tag
An analysis exploring the rationale behind continued investment in Intel for AI, despite competition and challenges.
CAT-Q introduces a post-training ternary quantization method for LLMs that uses learnable modulation and softened ternarization, achieving superior performance over BitNet 1.58-bit while using only 512 calibration samples and scaling to 235B parameters.
Valve is working with Intel and Nvidia to expand SteamOS support to more GPUs and handhelds, with initial firmware for Intel handhelds and ongoing driver work for Nvidia.
Die analysis of the Intel 8087 math coprocessor's fast bit shifter, exploring its architecture and role in floating-point operations.
A retrospective analysis argues that the 8086 segmented memory architecture was a clever design that could have scaled gracefully, but software developers' insistence on treating memory as a flat space led to its perceived flaws.
A user questions why AutoRound, a quantization tool offering superior accuracy retention at low bits and direct GGUF export, is overlooked despite outperforming standard AWQ and RTN, especially on complex models like Qwen3.6 27B.
Former President Donald Trump claims Apple will partner with Intel for chip design and production in the US, signaling a potential shift in Apple's supply chain strategy.
A detailed reverse-engineering analysis of the 69-bit adder in Intel's 8087 floating-point coprocessor from 1980, explaining its fast carry techniques and role in computing transcendental functions.
An interview with Kira Boyko, Intel's Xeon 6+ product director, discussing her role in defining the product, SKU strategy, and customer engagement, with the product launched at Computex 2026.
NVIDIA has released a 550B parameter model (Nemotron-3-Ultra-550B), prompting commentary that AMD and Intel should similarly release their own AI models as model development becomes a commodity for hardware companies.
Foxconn and Intel are partnering to develop next-generation AI systems, combining Foxconn's manufacturing capabilities with Intel's chip technology.
A tweet recounts Intel's historic 25% stock crash to an 11-year low, when the CEO posted prayers, contrasting with its current $500 billion valuation and hype edits.
The Verge's Sean Hollister previews the MSI Claw 8 EX AI Plus handheld with Intel's new Arc G3 Extreme chip, claiming significant performance and battery life improvements over AMD's flagship, potentially making Intel the leader in portable gaming chips.
This article describes how to use the SYCL backend with llama.cpp to achieve over 60 tokens per second on the Qwen 3.6-35B-A3B model using an Intel Arc Pro B70 GPU, with the entire model and KV cache in VRAM.
Jensen Huang made many jokes at the ARM press conference, including interactions with the Arm CEO, a signed notebook boomerang, self-deprecating about getting older, and shouting about buying ARM before leaving.
Benchmark results for Intel Arc Pro B70 GPU running llama.cpp with SYCL on Qwen models show 63 tokens per second performance.
Intel launched the Crescent Island GPU at Computex 2026, featuring up to 480GB VRAM and based on the Arc Xe 3P architecture, targeting next-generation AI workloads.
Intel announces its upcoming Crescent Island AI chip, designed for inference tasks, which will be cheaper and air-cooled compared to Nvidia and AMD offerings, aiming to ship by end of year.
Dell announces a new XPS 13 as a budget-friendly competitor to the MacBook Neo, launching at a promotional student price of $599 (until September) and later $699. The laptop features a thin 0.5-inch design, 2.2-pound weight, 13.4-inch anti-glare touchscreen, and an Intel Core 5 320 'Wildcat Lake' chip with 8GB RAM and 512GB storage.
Intel's upcoming Panther Lake processor is positioned to compete in local AI with up to 180 TOPS, surpassing AMD's Ryzen AI 400, amid the Microsoft-Nvidia-Windows-Arm N1X discussions.