NVIDIA H100, every mention
22 scenes · ← back to NVIDIA H100
tap a year for its mentions
every year anyone Brad Gerstner 22Dylan Patel 3Satya Nadella 2Sundeep Madra 1Clark Tang 1
Verbatim, from the transcripts: the passages where NVIDIA H100 comes up
The SpaceX IPO, Fable 5, AI Capex Update & Market Check w/ Gavin Baker, Andrew Fox & Clark Tang · Bg2 Pod
- ▶ 32:38 Clark Tang They were, you know, they, they didn't have that many H-One hundreds.
NVIDIA: OpenAI, Future of Compute, and the American Dream | BG2 w/ Bill Gurley and Brad Gerstner · Bg2 Pod
- ▶ 57:07 Brad Gerstner And we talked about Colossus I and what he, you know, what he was doing there, standing up a couple 100,000, you know, at the time, H-one hundreds, H-two hundreds in a coherent cluster. 2 times in the scene
China Open-Source, Compute Arms Race, Reordering Global Trade | BG2 w/ Bill Gurley and Brad Gerstner · Bg2 Pod
- ▶ 27:03 Brad Gerstner If you look at this one out of Elon talking about the x.ai goal is fifty million in units of H-one hundred equivalent, and Clark Tang on my team tweeted something that broke that down, which showed that that reflected something like
Stargate, Executive Orders, TikTok, DOGE, Public Valuations | BG2 w/ Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 16:28 Brad Gerstner That those GPUs, which are much more powerful, Blackwell, um, and eventually Rubin GPUs will be the equivalent, um, to millions of H-one hundreds.
AI Semiconductor Landscape feat. Dylan Patel | BG2 w/ Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 1:04:00 Dylan Patel It's a commodity in that sense, but because reasoning requires so much more memory, and the cost of goods sold of an H 100 to Blackwell, the percentage of cost to HBM has grown faster than the percentage of cost to leading edge silicon,…
- ▶ 1:13:10 Dylan Patel But like, they'll go down because they, you know, and, and, and then you're like, oh, well, like what's the market clearing price for, uh, if I wanted an H 100 for two years or a year.
- ▶ 1:22:23 Dylan Patel The problem is, nowadays, if you look at rental prices for H-one hundreds, they're tanking, right?
Satya Nadella | BG2 w/ Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 43:05 Satya Nadella And again, the fact that we've had two years of runway on it, uh, I think I like that business a lot more, and that's one of the reasons why the adverse selection problems here would have been lots of tech startups all looking for their… 2 times in the scene
Ep18. Jensen Recap - Competitive Moat, X.AI, Smart Assistant | BG2 w/ Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 27:50 Brad Gerstner And Larry Ellison apparently had, and he's like, you know, just, just because that dinner occurred and they ended up with a 100,000 H 100 don't necessarily, you know, connect the dots.
Ep17. Welcome Jensen Huang | BG2 w/ Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 46:57 Brad Gerstner They talked to you out of a 100,000 H-one hundreds.
Ep9. GPT-4o, Astra, Multi Modal, China Tariffs, Tech Earnings | BG2 with Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 26:26 Brad Gerstner We've talked on this pod and, and others about how Inflection, who was building, right, remember Pi, which stood for personal intelligence, um, was trying to build a consumer app, built one of the largest H 100 clusters in, in the world,…
Ep8. AI Models, Data Scaling, Enterprise & Personal AI | BG2 with Bill Gurley & Brad Gerstner · Bg2 Pod
- ▶ 52:13 Sundeep Madra But it, it, it intersects with that really big purchase of, um, you know, H-one hundreds that they talked about as well.
Ep6. AI Demand / Supply - Models, Agents, the $2T Compute Build Out, Need for More Nuclear & More · Bg2 Pod
- ▶ 11:11 Brad Gerstner If you are trying to build your own LLM today as inflection was, and you want to get your hands on 40,000 H 100, which I think they did, that's, that's like,
Ep5. Stock Comp, AI Cold War, Valuations for LLM | BG2 with Bill Gurley, Brad Gerstner, & Bob Mylod · Bg2 Pod
- ▶ 1:12:47 Brad Gerstner Well, inflection, you know, is, um, you know, just as a reminder, um, was one of the early, um, uh, LLMs to be trained on H-one hundreds.
Ep4. Tesla FSD 12, Imitation AI Models, Open vs Closed AI Models, Delaware vs Elon, & Market Update · Bg2 Pod
- ▶ 22:03 Brad Gerstner Clearly in the lead, they have bigger H-
- ▶ 29:29 Brad Gerstner Um, and then we have, you know, um, those models were trained on a mixture, I think, of H 100 and prior version of Nvidia chips. 4 times in the scene
- ▶ 48:49 Brad Gerstner And, you know, the fact of the matter is you need to have a million H 100.
BG2 w/ Bill Gurley & Brad Gerstner | NVDA, Chips, AI Compute Build Out, AI Impact on Big Tech | E03 · Bg2 Pod
- ▶ 9:25 Brad Gerstner So the COWAS data, you know, the order books, the H 100 data, and I think what people are seeing, and there's been some tweets about this, is that the lead times on NVIDIA H 100 are going down. 2 times in the scene
- ▶ 1:02:53 Brad Gerstner And, you know, is order of magnitude better than the H 100 that are out there today.
- ▶ 1:15:34 Brad Gerstner What I'm certain of, you know, what I, the rumored pricing of H 100 to B 100 is that B 100 are going to cost even more than the H 100. 2 times in the scene
BG2 with Bill Gurley & Brad Gerstner | MANG VC Gone Wild, Can You Trust AI Valuations? & More | E01 · Bg2 Pod
- ▶ 22:21 Brad Gerstner You saw that tweet or thread out of Zuckerberg this week, um, where he said by the end of the year, they're going to have the equivalent or this year, 600,000 H 100 GPUs, uh, running in super compute and that he was, and there's been a lot…
- ▶ 29:07 Brad Gerstner Well, because if you want to train the, you know, on 600,000 H 100, like we just talked Zuckerberg doing with the Lama three model, you have to have a business that is generating the massive cash piles that he's generating in order to do…