Nvidia Blackwell, every mention
18 scenes, the whole family · ← back to Nvidia Blackwell
tap a year for its mentions
every year anyone Batuhan Taskaya 7Alessio Fanelli 6Philip Kiely 4Shawn Wang 3Micah Hill-Smith 2Kyle Kranen 2Justin Johnson 2Ali Taha 2Thomas Sohmers 1Eiso Kant 1
Verbatim, from the transcripts: the passages where Nvidia Blackwell comes up
Next 100x in AI: Inference, Networking, & Self-Optimizing Models — Philip Kiely & Ali Taha, Baseten
- ▶ 14:49 Philip Kiely You're taking the model from, generally, these models are not released in NVFP four, um, and we want them to be in NVFP four for maximum blackwell compatibility, so we have to perform that quantization, um, and, you know, calibrate the…
- ▶ 39:11 Alessio Fanelli If you break down the two to four X, say, say the example is run GLM five two on B 200, single node, right? 3 times in the scene
- ▶ 49:49 Ali Taha And it took off and it was implemented on local devices because your, your memory bandwidth is so slow on like a MacBook, for instance, but try putting the same thing on like an Nvidia GPU on a B 200. 2 times in the scene
- ▶ 57:08 Philip Kiely That's, that's unsurprising.
- ▶ 1:12:35 Alessio Fanelli With the Rubens, you now have, what, NVL-SVII rack of plenty terabytes? 3 times in the scene
- ▶ 1:12:41 Philip Kiely Yeah, now, you, you still have NVL-SVII on, on, uh, Blackwell as well, but, um, you, you can't necessarily assume you're gonna do inference on that. 2 times in the scene
The AI Frontier: from open weights to open research — Eiso Kant, Poolside AI
Agent Inference at the "Speed of Light" — How NVIDIA moves like a $4.3 Trillion Startup
- ▶ 1:13:23 Kyle Kranen The Blackwell, like, RTX, 6000 cards. 2 times in the scene
Artificial Analysis: The Independent LLM Analysis House — with George Cameron and Micah Hill-Smith
- ▶ 1:03:02 Micah Hill-Smith When you run all of that, for especially big sparse models, you can get a lot better than two or three eggs gain going from hopper to blackwell generation to video. 2 times in the scene
After LLMs: Spatial Intelligence and World Models — Fei-Fei Li & Justin Johnson, World Labs
- ▶ 13:01 Justin Johnson Like, if you look at, if you look at the numbers, like, even going from Hopper to Blackwell, like, the performance per watt is about the same. 2 times in the scene
⚡️Raising $1.1b to build the fastest LLM Chips on Earth — Andrew Feldman, Cerebras
- ▶ 3:09 Andrew Feldman You think 20 times faster than Nvidia B 200 GPUs and it's, it's been an amazing run.
A Technical History of Generative Media
- ▶ 15:25 Batuhan Taskaya Uh, when you use PyTorch with B-Torch's Blackwell chips, you're not getting the best performance.
- ▶ 24:03 Batuhan Taskaya Blackwell is, is obvious, like, we, we have, like, we have five people dedicated to writing Blackwell kernels right now to make sure we can, like, because theoretically it looks good, right? 6 times in the scene
⚡️Accelerators @ 3x NVIDIA H200 perf, Made in the USA - Thomas Sohmers + Mitesh Agrawal, Positron AI
- ▶ 14:53 Thomas Sohmers Percentage of theoretical memory bandwidth actually has gotten worse generation over generation, and we'll see exactly where Blackwell ends up, but all indications are, even though they, you know, more than doubled the theoretical memory…
⚡️Math Olympiad gold medalist explains OpenAI and Google DeepMind IMO Gold Performances
- ▶ 30:46 Dr. Jasper Zhang Now you get B. 200 is faster.
The Shape of Compute (Chris Lattner of Modular)
- ▶ 5:27 Chris Lattner And as you do that in Blackwell, like all this stuff is like all now in the product.
2024 Year in Review: The Big Scaling Debate, the Four Wars of AI, Top Themes and the Rise of Agents
- ▶ 1:11:11 Shawn Wang I think the quote that I highlighted in AI news was that it is the best, like Blackwell is the best selling series in the history of the company. 2 times in the scene
- ▶ 1:22:00 Shawn Wang And obviously I think a lot of people are speculating how much more next year will be with H 200 becoming commodity blackwells coming out.