Nvidia, every mention
160 scenes, the whole family · ← back to Nvidia
tap a year for its mentions
every year anyone Matt Turck 120Dylan Patel 33Bryan Catanzaro 32Stephen Balaban 21Dan Fu 21Nathan Benaich 12Sharon Zhou 10Gaurav Misra 8Benedict Evans 5Andrew Feldman 5
Verbatim, from the transcripts: the passages where Nvidia comes up
When AI Improves Itself | Richard Socher (Recursive)
- ▶ 1:08:59 Richard Socher Um, uh, which is very useful for all, like, large, uh, hyperscalers and people who provide tokens, uh, and run models, uh, and we're very excited to keep pushing those, and we've heard very positive, uh, feedback from, from folks who are…
AI Could Take Over in 2029. Is It Already Too Late?
- ▶ 40:22 Ryan Greenblatt So if hypothetically it was the case that tomorrow a recipe for training super, super intelligence on like, you know, 64 H 100 or like some small amount of compute just dropped.
“OpenAI’s Model Hacked Us” - Hugging Face’s Thomas Wolf
- ▶ 36:20 Thomas Wolf But yeah, this void was filled rather quickly by companies like Reflection, uh, Thinking Machine, RC, uh, Mistral is supposed to open source a model soon, Nvidia themselves,
- ▶ 38:21 Thomas Wolf So for instance, the model we use to counter, um, open AI intrusion was GNM 5.2 that was quantized by Nvidia in, in, in four bits.
- ▶ 43:41 Matt Turck But if, if you think of the West, then, so you mentioned Nvidia and we had, uh, Brian Catanzaro, um, uh, from, you know, the whole NemoTron effort on the podcast uh, a few weeks ago. 2 times in the scene
- ▶ 51:59 Matt Turck So I think four days after the open wage letter, um, that, um, you know, NVIDIA did that we're talking about a minute ago, there was a different letter that came out with, uh, 1100 people.
Cerebras CEO: Why GPUs Can't Do Fast Inference
- ▶ 1:02 Matt Turck We started from what is a wafer and built up step by step, why GPUs struggle with fast inference, the three shortages nobody talks about, the decade in the desert when nobody wanted this chip, and why Andrew believes that CUDA is no longer…
- ▶ 5:38 Andrew Feldman And today, several companies make traditional GPUs, so NVIDIA and AMD, they make very standard GPUs. 4 times in the scene
- ▶ 13:58 Matt Turck NVIDIA had some announcement around just building chips for Windows computers.
- ▶ 15:29 Matt Turck And NVIDIA lost three hundred billion that day. 2 times in the scene
- ▶ 30:39 Andrew Feldman Our strategy was that you will never beat a great company like NVIDIA by doing something a little bit behind the feet.
- ▶ 1:01:06 Matt Turck So, you know, famously, uh, NVIDIA has, uh,
OpenAI’s Compute Chief: We Can’t Build Fast Enough | Sachin Katti
- ▶ 31:48 Sachin Katti That's a very big, uh, GB black belt cluster, uh, for our needs.
Inside Nemotron & NVIDIA’s AI Lab | Bryan Catanzaro
- ▶ 0:43 Matt Turck Brian Catanzaro leads NemoTron, NVIDIA's family of open foundation models. 2 times in the scene
- ▶ 1:39 Matt Turck So you guys at NVIDIA just released Nemo
- ▶ 4:25 Matt Turck Is that the communities and big companies like Nvidia being behind it? 2 times in the scene
- ▶ 10:10 Bryan Catanzaro Um, and you know, we're pushing Nemo Tron along here at Nvidia as well.
- ▶ 12:58 Bryan Catanzaro So, um, I started work at, uh, NVIDIA in 2008. 6 times in the scene
- ▶ 17:58 Bryan Catanzaro I, you know, I've always loved Nvidia. 7 times in the scene
- ▶ 22:09 Matt Turck NVIDIA care in the first place to be building model and investing very significant efforts into creating its own family of frontier models. 10 times in the scene
- ▶ 29:32 Bryan Catanzaro Um, more important to us than any particular generation is just the sustained commitment, ah, that NVIDIA has to developing these models. 3 times in the scene
- ▶ 30:28 Bryan Catanzaro We are increasing our investments substantially, and NVIDIA is a company that follows through. 5 times in the scene
- ▶ 34:42 Bryan Catanzaro But, um, you know, uh, the most important thing from NVIDIA's point of view that people are doing with, uh, uh, with LLMs is agents, right, is, um, building agentic workflows 2 times in the scene
- ▶ 44:37 Bryan Catanzaro So with Blackwell, for example, NVIDIA went all in on MOEs.
- ▶ 44:37 Bryan Catanzaro So with Blackwell, for example, NVIDIA went all in on MOEs. 4 times in the scene
- ▶ 52:17 Bryan Catanzaro And what that implies is that, you know, if we're trying as NVIDIA, as a company, to provide meaningful acceleration to the world's most important computational workloads, this has to be an important part of how we think about it. 2 times in the scene
- ▶ 1:00:23 Matt Turck So just taking a step back, like, tell us about the research organization at NVIDIA. 7 times in the scene
- ▶ 1:02:00 Bryan Catanzaro We had the Nemo team, which was building software for AI, and the Megatron team, which was building primarily focused on systems research for, um, for building large language models, and, you know, we decided to work together and then…
- ▶ 1:04:04 Matt Turck But you, you mentioned, uh, earlier that, um, despite the fact that, uh, you work at the, you know, number one undisputed, uh, leader in, in GPUs, uh, you all as a research organization don't have all the GPUs, uh, that, that you would… 2 times in the scene
- ▶ 1:08:34 Matt Turck Is that how the moonshots at NVIDIA got started as well over the years, whether that's in, in AI or otherwise? 3 times in the scene
- ▶ 1:10:54 Matt Turck Listening to everything, uh, you, you saying, I'm struck by how entrepreneurial the culture at Nvidia still seems to be. 8 times in the scene
Cloudflare CEO: The Internet's Business Model Is Dead
- ▶ 41:15 Matthew Prince Um, and that graphics chip company was NVIDIA and, um,
The GPU Myth: State of AI Compute 2026 | Stephen Balaban
- ▶ 0:54 Matt Turck How you finance billions of dollars of data centers and chips, why a 2023 H-one hundred can be more expensive to list today than when it was bought, and what it actually takes to stand up a gigawatt scale AI factory.
- ▶ 2:57 Stephen Balaban Yeah, so when you look at, for example, I think it's actually worth doing is to try to, like, kind of dig into some of the methodology on, for example, an index like, there's, there's the, there's the index that's on Bloomberg for H 100…
- ▶ 13:01 Stephen Balaban Every single modern deployment of, let's say, a Blackwell class or a Rubin class GPU, you know, the VR, GBN VR GPUs.
- ▶ 13:01 Stephen Balaban Every single modern deployment of, let's say, a Blackwell class or a Rubin class GPU, you know, the VR, GBN VR GPUs.
- ▶ 17:24 Stephen Balaban If you look at the cost structure of let's say one GPU hour of time, you know, we were talking about H 100.
- ▶ 19:44 Stephen Balaban You know, let's say NVIDIA, GB, 300 and VL, 72 racks.
- ▶ 24:59 Stephen Balaban Um, if you were to kind of break down NVIDIA's, uh, bill of materials, then, you know, you can kind of get better allocation towards, uh, where those costs are coming from. 2 times in the scene
- ▶ 25:22 Matt Turck So you guys are a big Nvidia shop at a precise level. 8 times in the scene
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 25:53 Stephen Balaban We've got everything from V-one-hundredths, A-one-hundredths, H-one-hundredths, H-two-hundredths, uh, B-two-hundredths, G-H-two-hundredths, or GB-two-hundredths, B-three-hundredths, and VR-two-hundredths coming soon.
- ▶ 40:28 Stephen Balaban Um, but it's becoming more and more mature, and in general, creditors and lenders are really starting to understand the value of an NVIDIA chip.
- ▶ 40:40 Stephen Balaban Because, you know, you actually look at the chips that we deployed in twenty-twenty-three, H-one hundreds. 2 times in the scene
- ▶ 43:07 Stephen Balaban Um, I'm starting to see that, but fundamentally, I think that the, the, the asset class is just starting to mature and creditors are starting to become very comfortable with in investing in the credit side of buying Nvidia GPUs and… 2 times in the scene
Benedict Evans: OpenAI’s Moat Problem & the Future of Software
- ▶ 40:05 Benedict Evans The same with NVIDIA. 3 times in the scene
Mistral AI vs. Silicon Valley: The Rise of Sovereign AI
- ▶ 3:50 Matt Turck Uh, so that was a big announcement, uh, I guess in June of 25, putting a big partnership with NVIDIA to, um, help with this effort.
- ▶ 9:20 Matt Turck So you have a bunch of partnerships with, um, SAP and NVIDIA, uh, but there is, you, you don't have one of those gigantic companies, uh, on your cap table.
- ▶ 41:37 Timothée LeCroix Definitely compute and, uh, the current deployment that we have will help, uh, as it's going to be giving us a lot more Grace Blackwell capacity than we had in the past.
Dylan Patel: NVIDIA's New Moat & Why China is "Semiconductor Pilled”
- ▶ 0:43 Matt Turck Nvidia's massive move to acquire Grok, the truth about the CapEx bubble, whether the US power grid can actually handle the AI boom, and the geopolitical chess match playing out between the US and China.
- ▶ 1:17 Matt Turck I'd love to start with GROC and NVIDIA since it's still fresh. 8 times in the scene
- ▶ 3:37 Dylan Patel That's the CPX. 3 times in the scene
- ▶ 6:24 Dylan Patel I was an advisor for NVIDIA acquired in Fabrica just maybe a few months before they did Grok and similar style of deal, right? 3 times in the scene
- ▶ 7:00 Matt Turck So now is, uh, Nvidia also dominating the, the inference market. 8 times in the scene
- ▶ 10:25 Dylan Patel It's an incredible amount of the software that NVIDIA GPUs run on is not from NVIDIA. 6 times in the scene
- ▶ 17:16 Dylan Patel Because Blackwell is just way better than MI-tri-fifty-five.
- ▶ 17:19 Dylan Patel Um, and then, you know, Rubin comes out and they'll be way, way behind, but then AMD's new chip comes out and AMD will be caught up, or even slightly ahead.
- ▶ 17:40 Dylan Patel Yeah, I mean, NVIDIA's revenue this year is going to be like, it's a lot.
- ▶ 18:08 Dylan Patel You, you have to specialize because you're never going to beat Nvidia at their own game, right? 7 times in the scene
- ▶ 19:27 Dylan Patel It's not just Hopper, Blackwell, Rubin.
- ▶ 19:27 Dylan Patel It's not just Hopper, Blackwell, Rubin. 2 times in the scene
- ▶ 19:27 Dylan Patel It's not just Hopper, Blackwell, Rubin.
- ▶ 19:29 Dylan Patel Now it's going to be, you know, it's not Ampere Hopper, you know, you know, it's not that line.
- ▶ 19:41 Dylan Patel And, and actually AI models have developed in a way where CPX or Grok style chips are the best, right?
- ▶ 21:51 Dylan Patel Like, like if, for example, a chip optimized for video and image generation existed today and it was better than NVIDIA or NVIDIA made it, I think mid journey would absolutely only use that for inference.
- ▶ 22:57 Matt Turck Huawei and Nvidia in China, last year there was like, 10 or 12% of their overall revenue and 2 times in the scene
- ▶ 27:26 Dylan Patel But as far as I understand, the national government has not banned Nvidia's H-twenty or H-two hundred, but the local ones have.
- ▶ 27:26 Dylan Patel But as far as I understand, the national government has not banned Nvidia's H-twenty or H-two hundred, but the local ones have.
- ▶ 27:26 Dylan Patel But as far as I understand, the national government has not banned Nvidia's H-twenty or H-two hundred, but the local ones have.
- ▶ 34:31 Matt Turck Yeah, the two billion pipe that NVIDIA announced. 6 times in the scene
- ▶ 36:23 Matt Turck I mean, obviously the U.S. and China are the two biggest markets, but like for other markets, I don't know, UAE, Middle East, uh, Europe, are NVIDIA and Huawei already, uh, head to head in deals? 7 times in the scene
- ▶ 43:17 Dylan Patel TSMC is literally making chips for NVIDIA and Apple and AMD and others in Arizona today, right?
- ▶ 1:03:41 Matt Turck I wanted to come back, uh, quickly to, uh, that, um, NVIDIA and CoreWeave deal that you mentioned as we sort of, uh, uh, close the discussion on, uh, on CapEx and a, and a bubble. 4 times in the scene
State of LLMs 2026: RLVR, GRPO, Inference Scaling — Sebastian Raschka
- ▶ 33:02 Sebastian Raschka Just like last week, Nvidia also had a paper on GDPO, I think. 2 times in the scene
The End of GPU Scaling? Compute & The Agent Era — Tim Dettmers (Ai2) & Dan Fu (Together AI)
- ▶ 3:25 Dan Fu Um, on NVIDIA's Blackwell GPU.
- ▶ 3:25 Dan Fu Um, on NVIDIA's Blackwell GPU.
- ▶ 17:34 Dan Fu On last generation, kind of nerfed GPUs, H 800 instead of H 100, the 800 is nerfed by all sorts of ways from, from NVIDIA, um, to, to get around the, the expert restrictions at the time. 2 times in the scene
- ▶ 17:34 Dan Fu On last generation, kind of nerfed GPUs, H 800 instead of H 100, the 800 is nerfed by all sorts of ways from, from NVIDIA, um, to, to get around the, the expert restrictions at the time. 2 times in the scene
- ▶ 17:34 Dan Fu On last generation, kind of nerfed GPUs, H 800 instead of H 100, the 800 is nerfed by all sorts of ways from, from NVIDIA, um, to, to get around the, the expert restrictions at the time. 2 times in the scene
- ▶ 19:15 Dan Fu They're building out tens of thousands of, of B-two hundred, GB-two hundred chips. 2 times in the scene
- ▶ 19:15 Dan Fu They're building out tens of thousands of, of B-two hundred, GB-two hundred chips. 4 times in the scene
- ▶ 21:06 Dan Fu Hey, this was trained on H 100, H 200 and GB 200, which to me suggests
- ▶ 29:58 Matt Turck You know, obviously there's been Grok and NVIDIA, there's, uh, Cerebras, there's like a bunch of, uh, sort of special ASICs companies coming up, uh, from, from your kind of like low level in the stack, uh, Vantage Point. 6 times in the scene
- ▶ 56:23 Dan Fu And the way that we've been writing kernels for the whole history of, let's say call it NVIDIA hardware, is that you really specialize a single kernel for a single operation. 2 times in the scene
- ▶ 59:41 Tim Dettmers you can fit it pretty well, even sort of low-grade data center GPU, like an RTX 6000, cost 6000 dollars.
- ▶ 1:01:15 Dan Fu So we're starting to hear a little bit about Rubin.
- ▶ 1:01:18 Dan Fu Uh, the next generation of NVIDIA GPUs.
- ▶ 1:02:33 Dan Fu I think Nvidia released a bunch of really great hybrid models recently. 2 times in the scene
AI Eats the World: Benedict Evans on What Really Matters Now
- ▶ 31:04 Benedict Evans NVIDIA to me wants to be the new Sun. 2 times in the scene
Dashboards Are Dead: Sigma’s BI Revolution for Trillion-Row Data
- ▶ 7:35 Mike Palmer So if you're familiar with Sutter Hill Ventures, and Sutter Hill Ventures, uh, is a pretty well-known, uh, uh, VC in the valley and behind companies like Nvidia, for example, but certainly the, uh, Snowflake.
Why This Ex-Meta Leader is Rethinking AI Infrastructure | Lin Qiao, CEO, Fireworks AI
- ▶ 45:54 Matt Turck Like, do you run on like all sorts of NVIDIAs and MDs and like, are you agnostic that way? 2 times in the scene
From Selfie to Studio: Captions CEO on AI Video for 10M Creators
- ▶ 48:48 Gaurav Misra And by the way, like the H 100 and A 100 kind of suck at encoding and decoding video, right? 2 times in the scene
- ▶ 48:48 Gaurav Misra And by the way, like the H 100 and A 100 kind of suck at encoding and decoding video, right? 3 times in the scene
page 1 of 2 · 100 scenes per page · newest episode first next →