viraldi.
Back to discovery
29:48
Engineering & Industry / VIDEO

It’s von Neumann’s fault | GPU · TPU · LPU

Julia Turc· 83.2K subscribers

2d ago · en-US audio

In this video, we discuss the core differences between the dominant AI inference chips: NVIDIA GPUs, Google TPUs and Groq LPUs. We cover core concepts including the von Neumann bottleneck, systolic arrays and static scheduling. The video also features snippets from interviews with Jonathan Ross (founder and CEO of Groq, now at NVIDIA) and Abdel Sghiouar (senior developer advocate at Google Cloud). Resources: - TPU…

3.2×Score at discovery
121.7KViews
37.6KMedian · 13 videos

Outlier score

This score was measured on Thu, 05 Nov 2026 03:17:56 GMT. Current view counts may differ.

119.7K views ÷ a 37.6K baseline = 3.2×. Compared with 13 earlier videos from the same channel, matching the format recorded at scoring time. Older-upload baseline: reference videos had at least as long to gain views. This comparison is not age-matched.

YouTube snapshot: Thu, 05 Nov 2026 06:12:15 GMT. Counters may have changed.

Read the methodology

Niches & tags

Classification is incomplete; the primary niche may change.

Transcripts and detailed analytics are available for recent outlier Shorts up to three minutes. Research tools