#Ai Chips
All ObjectWire articles tagged with “Ai Chips”.
3 articles
Tech
Meta and Broadcom Extend Custom AI Chip Partnership to 2029 | $2.3B Paid, 2nm MTIA 500 Confirmed
The deal covers four MTIA chip generations, a one-gigawatt compute deployment, and a first-of-its-kind 2-nanometer custom AI accelerator. Broadcom CEO Hock Tan will leave Meta's board and move to a dedicated advisory role.
Jack Brenen · Apr 19, 2026
Tech
DeepSeek Founder Says V4 Model to Launch in Late April on Huawei Chips
Liang Wenfeng confirmed the timeline in an internal communication after months of missed deadlines, as DeepSeek rewrites its model stack for Huawei Ascend and deliberately withholds early access from Nvidia and AMD
Jack Brennan · Apr 11, 2026
Tech
Nvidia B300 vs AMD MI300X vs Google TPU v6 | 2026 AI Chip Specs, Workloads, Cost Comparison
DGX B300 leads on raw PFLOPS, MI300X leads on single-GPU VRAM, TPU v6 Trillium leads on energy efficiency, and none of them are directly comparable without caveats
ObjectWire Technology Desk · Apr 1, 2026