AMD Announces Ryzen AI Halo, the Compact DGX Spark and Mac Mini Rival
…Even in scenarios that aren't as memory-intensive, AMD claims some performance gains. Qwen 3.6 35B model shows the Ryzen AI Halo 4% faster, and GLM 4.7 Flash 30B…
Search
All sources
wccftech.com
15
tomshardware.com
10
theregister.com
9
hothardware.com
6
tweaktown.com
5
engadget.com
4
techcrunch.com
3
techpowerup.com
3
theverge.com
2
club386.com
2
fudzilla.com
2
huggingface.co
2
Videos
More videosTop stories
FuriosaAI Ditches GPU Playbook For 2nm Broadcom-Built Inference Chip, Claims HBM4/E Bandwidth Beats Even The Most Efficient GPUs
1w agoAngry TSMC employees considering strikes, unionization over employee bonuses, report claims — company reportedly considering 15% payout cut to fund capex despite record revenues fuelled by AI surge
1w agoHuawei claims sanctions-busting breakthrough with 1.4nm-class chips by 2031, claims 55% higher transistor density — firm claims new LogicFolding chip architecture can bypass EUV restrictions, introduces 'Tau Scaling Law' to replace Moore's Law
1w agoVertu Is Back With a Folding Phone Powered by—Surprise—an AI Agent
…In late 2025, it unveiled the Agent Q, which it calls the “world's first AI agent phone for entrepreneurs.” While the company still claims a British heritage, its phones are no…
AMD Courts Gamers At Computex With New X3D CPUs, GPUs, And EXPO ULL RAM
…The performance claims are certainly attention-grabbing, with AMD citing up to 13 percent better average frame rates and 15 percent better one-percent lows on a Ryzen 7 9700X when compared…
Semiconductor News, Analysis and Features | Tom's Hardware
…Premium Premium TSMC claims that CoWoS innovations will enable 48x more compute and 34x more memory bandwidth for 2029 AI processors. Premium SPONSORED SPONSORED_STRAPLINE SPONSORED
Discussions and forums
New Google Gemma 4 12B Claims Near-26B Performance - We Tested Both!
We ran both models locally on one RTX 4090 and gave each the same task: write a self-contained HTML5 canvas animation with real physics in one file without libraries. Three scenes - a Galton board, two blocks colliding o…
Reality check from the Microsoft AI Tour: "Agents" hype, the enterprise disconnect, and peak AI Fatigue
Just got back from the Microsoft AI Tour in Zurich. Honestly? Nothing has globally changed since my last visit to these events two years ago. They just scrubbed "LLM" and "GenAI" from all the slides and replaced them wit…
Electronics Manufacturing News, Analysis and Features | Tom's Hardware
…Premium Premium TSMC claims that CoWoS innovations will enable 48x more compute and 34x more memory bandwidth for 2029 AI processors. Premium Sponsored SPONSORED_STRAPLINE SPONSORED
ASUS Unveils ROG NUC 16 Compact Gaming PC With RTX 5080
…ASUS claims the system can deliver up to 1334 AI TOPS of processing capability while offering up to 16 GB of dedicated GDDR7 graphics memory. The company positions the machine for local…
Nvidia eyes CPU crown – Fudzilla.com
…AI workloads. Nvidia said Vera was co-designed with Rubin GPUs and NVLink to deliver up to 1.5 times faster per-core performance. It claims Vera delivers twice the performance per…
AMD Claims Leadership Tokens/$ With Its Ryzen AI Halo Dev Platform, Tackles NVIDIA's Spark at $3999 & Pays For Itself Within 6-Months
…These high-performance and premium SoCs offer amazing performance thanks to their Zen 5 CPU, RDNA 3.5 GPU, and XDNA 2 NPU architectures. The Ryzen AI Halo combines these SoCs in…
After Gobbling Up DRAM, NVIDIA & SK hynix Plan to Introduce an “AI SSD” With 10× Higher Performance, Ringing Alarms Over NAND Supply
…Read all comments on After Gobbling Up DRAM, NVIDIA & SK hynix Plan to Introduce an “AI SSD” With 10× Higher Performance, Ringing Alarms Over NAND Supply
Nvidia crams 256 Vera CPUs into a single liquid cooled rack
…The company is making some rather bold claims about its latest CPU superchips. If Nvidia is to be believed, Vera will deliver 3x more memory bandwidth and 1.5x the performance per…