Ep. 2 - Bryan Catanzaro on Nvidia's open models and risky bets

from localhost · host Atila

Atila Orhon, founder of Argmax, locally hosts Bryan Catanzaro, co-creator of cuDNN, member of the Deep Speech Mafia and VP of Applied Deep Learning at Nvidia:(00:20) Who is Bryan?(07:38) Getting Nvidia to care about Deep Learning(14:13) Why did Bryan leave Nvidia right when Deep Learning was taking off(18:02) Leadership: Aligning a village of researchers(24:12) Will the frontier flip back to open? (32:16) Nvidia's models: Side project or core business? (38:19) Efficiency leads to edge inference: Does Apple capture inference?(42:43) Nvidia’s risky bets: Fewer and fewer bits(47:19) Nvidia's misstep with Volta(52:30) Every model is already obsolete as soon as you stop training it

Episode metadata supplied by the publisher feed · Published Apr 10, 2026

Embed this episode

Attribution link and audio player

Ready to play

Ep. 2 - Bryan Catanzaro on Nvidia's open models and risky bets

0:00 58:42

1×

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

Share this episode

Similar Episodes

No similar episodes found.

Similar Podcasts

Localhost Podcast Localhost Podcast Hello from the Internet!This is your guide to the various aspects of the lovely world of web development.Every month we shall get together, and discuss an aspect of web development. We will unwrap the subject using questions sent by YOU the listener. You can contact us via the twitter on @localhostfm or email your questions to [email protected] Drew and Rob Dudley are stepping up to the challenge to answer these questions for you. Localhost Andreas Donner Localhost ist der Netzwerk-Podcast. Hier spricht Andreas Donner, Chefredakteur von IP-Insider, mit Experten und Anwendern aus der Netzwerk-Welt. Webmestre.fm Kevin Huddy et Francis Parent-Valiquette Programmeur web "freelance", ce podcast est votre nouveau localhost/public_html! Francis et Kevin (2 programmeurs web old-school) se rencontre afin de partager trucs et astuces pour gagner du temps, servir mieux vos clients et faire plus d'argent! the localhost the localhost Five tech industry veterans argue about AI that runs on your own hardware. Open weight models, NPUs, home labs, and the real question underneath it all: who controls intelligence, you or a data center?Every week, one host takes the hot seat to defend a provocative claim while the other four try to take it apart. Then everyone votes. We run the models, test the vendor claims ourselves, and share what actually happened, including the failures.No hype, no scripts, and nobody here is pretending to have all the answers. We're learning this stuff alongside you, one episode at a time.New episodes weekly. There's no place like 127.0.0.1.

Frequently Asked Questions

How long is this episode of localhost?

This episode is 58 minutes long.

When was this localhost episode published?

This episode was published on April 10, 2026.

Can I download this localhost episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.

URL copied to clipboard!