Z.ai Says Its Cheapest Model Runs Entirely on Chinese Chips. Nobody Can Check.
Z.ai confirmed it built Ox Alpha, now rebranded GLM-5.3-Flash, and claims the model runs entirely on 100,000 domestically made chips. The company won't say…
Z.ai confirmed it built Ox Alpha, now rebranded GLM-5.3-Flash, and claims the model runs entirely on 100,000 domestically made chips. The company won't say…
Nvidia's AI server prices are set to increase by over 15 percent due to surging memory costs from suppliers like Samsung and SK Hynix.…
Seven. That’s how many generations of Google’s Tensor Processing Unit Amir Salek shipped before he left in 2022, and it matters more than his…
CME Group is launching two futures contracts on October 5 that will allow trading of Nvidia GPU rental prices, reflecting their high demand in…
In July, Moonshot AI faced accusations of misusing Nvidia chips, but it only rented compute from Thailand, exploiting a loophole in U.S. export controls.…
Mistral Large 3 shares DeepSeek V3's attention stack down to the compression ranks, and since 11 August Mistral serves Z.ai's GLM-5.2 unmodified on its…
Cerebras unveiled the CS-4, a powerful server rack featuring three wafer-scale chips, promising significant throughput improvements. However, despite reported advancements and a $180.1 million…
A Guardian investigation into internal Microsoft documents found the company has 2.2 million AI chips installed, a fraction of what its own gigawatt claims…
Firebird's Armenia AI factory, the biggest in the former Soviet bloc, is really a story about export licenses: Washington cleared Nvidia Blackwell into a…
The U.S. State Department's recent letter urges 35 countries to choose sides in the AI competition between Washington and Beijing, highlighting concerns over Kazakhstan's…
TSMC's 3nm production is set to reach 180,000 wafers monthly by early Q4, ahead of schedule due to high demand from Nvidia, AMD, and…
Nvidia is partnering with major private capital firms to arrange $500 billion in AI infrastructure financing, allowing customers to purchase its GPUs and build…
The article discusses the evolution of NVIDIA's AI technologies, particularly the Tensor Cores and their role in enhancing graphics rendering through techniques like DLSS…
The Firebird AI factory needs 300 megawatts in a country whose grid peaks near 1,300 and leans on an aging Russian-fuelled reactor and Russian…
Qualcomm's acquisition of Arduino aims to transform robotics by integrating edge AI with real-time control on a single board. The new Uno Q and…
One maker's tech digest, weekly. Chips, AI, gaming, EVs — no spam.