Comp Ai

Nvidia Vera Rubin NVL72 beats Blackwell by 7x in DeepSeek inference test: Inference tests on a 1.6-trillion parameter DeepSeek model show Nvidia's Vera Rubin NVL72 delivers up to seven times better token throughput per megawatt compared to Blackwell. This performance exceeds CEO Jensen Huang's previous claim of a threefold improvement for models ranging from 1T to 3T parameters.

Hardware • Techmeme

XPeng deploys Turing AI chips in humanoid robots to deliver 2,250 TOPS: XPeng is extending its self-developed Turing AI silicon from intelligent electric vehicles into humanoid robots as part of its Physical AI stack. The company's IRON humanoid robot configurations will feature approximately three on-device Turing chips, delivering a total of roughly 2,250 TOPS.

Robotics • Pandaily

Perplexity deploys GPT-6 Astra to manage production systems: Perplexity has deployed GPT-6 Astra to write communications, modify software, and monitor production systems. The company reports checking in on the AI's work much less frequently compared to its use of earlier models.

GPT-6 Astra • OpenAI Blog

China Mobile Cloud debuts GPU-neuromorphic inference system: China Mobile Cloud and its partners have introduced a domestic GPU plus neuromorphic heterogeneous mixed-inference system for large models. Tested on DeepSeek V4 Flash, the system reportedly delivers approximately twofold output and energy gains alongside a 40% reduction in operating expenses.

Hardware • Pandaily

Two-year study shows banning AI in classrooms leaves students with worst performance: A law professor's two-year study compared the effects of an AI ban, unguided AI use, and structured AI training on student performance. The student group banned from using AI finished last in both years of the study, disproving the researcher's initial assumption that unguided AI use would cause harm.

AI Applications • The Decoder