Topic
Best Gpu technology Podcast Episodes
Gpu technology is covered across 2 podcast episodes in our library — including The All-In Podcast. Each episode brings firsthand experience and deep research from leading practitioners in the field.
Below you'll find key insights, core concepts, and actionable advice aggregated from the top episodes — followed by a ranked list of the best gpu technology discussions to explore next.
Key Insights on Gpu technology
- 1.The CoreWeave CEO believes the GPU depreciation debate, especially regarding H100s, is "nonsense" primarily instigated by short sellers.
- 2.CoreWeave's average client contract for compute services is five years, with some extending to six years, indicating long-term utility of their GPUs.
- 3.Claims that GPUs become obsolete in as little as 16 months are directly contradicted by CoreWeave's contractual agreements with customers.
- 4.CoreWeave employs a six-year depreciation schedule for its GPUs, anticipating that the hardware will remain viable for even longer.
- 5.The CEO's simple metric for a GPU's value is customer willingness to pay for its use over several years, validating its extended shelf life.
- 6.The company's "success-based" model means infrastructure is acquired based on client demand and long-term commitments, not speculative short-term depreciation.
Top Episodes — Ranked by Insight (2)
The All-In Podcast
"It's Nonsense Being Pushed by Short Sellers." - CoreWeave CEO on the GPU Depreciation Debate
The CoreWeave CEO believes the GPU depreciation debate, especially regarding H100s, is "nonsense" primarily instigated by short sellers.
The All-In Podcast
Jensen Huang: Nvidia's Future, Physical AI, Rise of the Agent, Inference Explosion, AI PR Crisis
Nvidia has evolved from a GPU company into an "AI factory company," integrating Groq, CPUs, BlueField, and networking processors for a diversified computing strategy.
Episodes ranked by insight density — scored on key takeaways, concepts explained, and actionable advice. AI-generated summaries; listen to full episodes for complete context.
