Samsung Spills The Beans Of Its Next Flagship Chipset, The Exynos 2700, Claiming Its “Enhanced AI Performance” Can “Expand Market Share”
… Unfortunately, apart from mentioning increased AI performance, no other details were mentioned. …
Search
Alibaba Targets NVIDIA's Hopper With Zhenwu M890 AI Chip, Claiming 3x The H20 Performance, 144GB HBM3 & A Roadmap Through 2028
… Read all comments on Alibaba Targets NVIDIA’s Hopper With Zhenwu M890 AI Chip, Claiming 3x The H20 Performance, 144GB HBM3 & A Roadmap Through 2028
NVIDIA Loses Ground With AI Engineers as Cooling and Power Costs Push Hyperscalers Toward Custom ASICs, Evercore Warns
… Desire To Improve Economics Is Driving Engineers Towards Alternatives To NVIDIA, Says Firm As per Evercore, the shift to an "inference-led regime" from a "training-led regime" is "increasing focus on cost-per-token, ROI and TCO, which is accelerating hyperscaler interest in homegrown ASICs and alte… …
FuriosaAI Ditches GPU Playbook For 2nm Broadcom-Built Inference Chip, Claims HBM4/E Bandwidth Beats Even The Most Efficient GPUs
FuriosaAI and Broadcom have partnered to build a high-performance AI accelerator chip featuring next-gen HBM4/E memory. …
Low Latency Wide DRAM Adopts HBM’s Integrated Design To Enable On-Device AI In Smartphones; 1.5 Times Higher Bandwidth With Lower Temperatures Promised
… Related Story Aletheia Warns HBM Prices Will Double in 2027 as Memory Becomes AI’s Most Critical Component The rumor claims that LLW delivers 1.5 times the performance while reducing power consumption by 50 percent, though it’s not specified which memory standard these metrics are compared to, so w… …
DeepSeek Aims At Memory Shortage With Latest AI Model But Might Sacrifice Performance
… Read all comments on DeepSeek Aims At Memory Shortage With Latest AI Model But Might Sacrifice Performance
A Chinese AI Chip Startup Unveils a Roadmap That Suprisingly Pledges to Be On-Par with NVIDIA's Cutting-Edge Vera Rubin By 2027
… These claims certainly look fascinating, but before you factor in power constraints and thermal requirements, we won't go into that debate for now. The key bottleneck for Chinese AI chip startups is the lack of a semiconductor ecosystem that could put them on par with Western options. …
Bolt Graphics Tapes Out Zeus GPU Which It Claims Is 5x Faster Than NVIDIA's RTX 5090 In Path Tracing At Half The Power
… This time, Bolt Graphics is sharing updated performance figures in which it is claiming a 5x increase in path tracing performance versus the RTX 5090. …
Tensordyne's 3nm Napier AI Chip Promises 13x Higher Token Throughput Than Blackwell & Blazes Past Rubin With 1000 Tokens/s In Multi-Trillion Parameter Models
US-based AI company, Tensordyne, has announced the successful tape-out of its Napier chip , which it claims to demolish NVIDIA's Blackwell & Rubin chips with leading token throughput and efficiency. …
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