InfiniBand
product on 8 shows · 9 statements across 8 episodes · said 47 times in 23 episodes since 2023
Acquired 15 
Latent Space 12 
No Priors 5 
the a16z Podcast 5 
the MAD Podcast 3 
All-In 3 
20VC 3 
BG2 Pod 1
Mentions by year, every show
tap a year for its mentions
Acquired 15Latent Space 12the a16z Podcast 5No Priors 5All-In 320VC 3the MAD Podcast 3BG2 Pod 1
2026 7 mentions in 4 episodes 2 per episode
- The GPU Myth: State of AI Compute 2026 | Stephen Balaban
- Agent Inference at the "Speed of Light" — How NVIDIA moves like a $4.3 Trillion Startup
- How to Build Your Own Data Center & Why Every Startup Should Do It
- The Future of AI Infra: from Kubernetes to Agent Sandboxes — Akshat Bubna, Modal CTO
- every mention in 2026, scene by scene →
2025 8 mentions in 5 episodes 2 per episode
- SF Compute: Commoditizing Compute
- Steeve Morin: Why Google Will Win the AI Arms Race & OpenAI Will Not | E1262 · 20VC with Harry Stebbings
- "Is there an AI bubble?” Gavin Baker and David George
- Dylan Patel on the AI Chip Race - NVIDIA, Intel & the US Government vs. China
- No Priors Ep. 96 | With Modal CEO and Founder Erik Bernhardsson
- every mention in 2025, scene by scene →
2024 16 mentions in 9 episodes 2 per episode
- State of the Art: Training 70B LLMs on 10,000 H100 clusters
- New SEC Chair, Bitcoin, xAI Supercomputer, UnitedHealth CEO murder, with Gavin Baker & Joe Lonsdale
- No Priors Ep. 89 | With NVIDIA CEO Jensen Huang
- No Priors Ep. 95 | Best of 2024
- Ep17. Welcome Jensen Huang | BG2 w/ Bill Gurley & Brad Gerstner · Bg2 Pod
- The Quest for Community-Trained Open Source AI Models
- No Priors Ep. 77 | With Foundry CEO and Founder Jared Quincy Davis
- Renaissance Technologies (Audio) · Acquired
- 1 more episode that year, every mention in 2024 →
2023 16 mentions in 5 episodes 3 per episode
every mention on every show, scene by scene, with the transcript →
9 statements about InfiniBand, every show
Scaling beyond eight NVIDIA H100 GPUs requires inter-GPU communication over InfiniBand
“The maximum NVLink domain, domain for H 100 for most DGX H 100 Is eight GPUs, right? So if you scaled up past that, You're gonna have to figure out ways to handle the fact that now for the GPUs to communicate, you have to do it over InfiniBand, which is still …”
Conrad: Decentralized compute networks will never beat co-located InfiniBand clusters
“I just don't really think this is gonna ever be more efficient than a fully interconnected cluster with Infiniband, or, you know, whatever sort of next spec might be. Like, I could be completely wrong, but Speedolite is really hard to beat. And regardless of w…”
Huang: NVIDIA Bought Mellanox to Treat Data Center Networks as Compute Fabrics
“That, that's the reason why we bought Mellanox and started fusing InfiniBand and MVLink in such an aggressive way.”
Schmidt: DisTrO will eliminate data centers' InfiniBand dependency before edge AI dominates
“I think there's, you know, immediately coming out. What you'll see is the ability for even centralized actors who might have multiple data centers to now, like just use them in a more efficient way. Like just have N equals two, you know, like anything. And eac…”
Frankle: Porch pirates stole MosaicML InfiniBand cables twice before data center delivery
“Our InfiniBand cables getting stolen from the data center twice, like in boxes before they arrived at the data center, like, you know, porch pirate basically had stolen our InfiniBand cables back when those were hard to come by”
Frankle: Large-scale model training forces teams to debug the full infrastructure stack
“It's kind of impossible if you're doing training to not go all the way through the entire stack, regardless of what happens. Like somehow I'm still chatting with cloud providers about power contracts, even though the whole point of dealing with the cloud provi…”
Gilbert: Mellanox likely built its InfiniBand business on quant finance
“A huge amount of InfiniBand was used by high-frequency trading firms, and I don't know for sure, but I kind of think Mellanox built their business on Quant Finance.”
Modal Is Not Designed for Training Large Foundation Models
“When you get to like training, like very large foundational models, that that's a use case we don't support super well because that's very high IO, you know, you need to have like infinite band and all these things. And those are things we haven't supported ye…”
Horowitz: Nvidia's InfiniBand networking gives it a major AI advantage over AMD
“One of their biggest advantages is They have a, you know, Nvidia also owns a networking architecture called InfiniBand, and InfiniBand turns out to be, like, really, really important in AI you know, in, in terms of having the, kind of, GPUs being able to commu…”