AMD, every mention
39 scenes, the whole family · ← back to AMD
tap a year for its mentions
every year anyone Dylan Patel 14Quentin Anthony 12Shawn Wang 8Chris Lattner 7George Hotz 6Thomas Sohmers 5Mitesh Agrawal 2comfyanonymous (Comfy) 2Suhail Doshi 1Soumith Chintala 1
Verbatim, from the transcripts: the passages where AMD comes up
The Inference Frontier: from 100 to 10,000 tokens per second — Sean Lie, Cerebras CTO
- ▶ 32:00 unnamed speaker I think before getting into etched and super spicy chips, any takes on AMD and NVIDIA, what they're doing?
⚡️ Google's Open AI Strategy — Omar Sanseviero, Google DeepMind
- ▶ 3:42 Omar Sanseviero So for example, we work with Lama CPP, Olama, MLX, Hogan Faces, BLM, NVIDIA, AMD.
The $15B Physical AI Company: Simulation, Autonomy OS, Neural Sim, & 1K Engineers—Applied Intuition
- ▶ 6:49 Qasar Younis Nvidia does or what an AMD, but we just don't do chips.
Dylan Patel Explains the AI War While Cooking | In-Context Cooking
- ▶ 39:47 Dylan Patel Um, and it is really an innovator's dilemma in the sense that, like, you know, hey, why did, why did Intel and AMD not win in AIG fuse?
One Year of MCP — with David Soria Parria and AAIF leads from OpenAI, Goose, Linux Foundation
- ▶ 1:02:05 David Soria Parra Uh, HDMI, uh, HDMI forum does not want to allow the AMD to develop open source Linux drivers for HDMI 2.1.
How Zyphra went all-in on AMD + Why Devs feel faster with AI but are slower — with Quentin Anthony
- ▶ 0:54 unnamed speaker Uh, we want to talk about AMD. 3 times in the scene
- ▶ 2:09 unnamed speaker When did you decide to go to AMD? 7 times in the scene
- ▶ 2:21 Quentin Anthony So when I was working on PhD more heavily, I was working on Frontier, which is a supercomputer Oak Ridge National Lab, National Lab that's totally based on MI- two-fifty X, uh, AMD GPUs. 3 times in the scene
- ▶ 3:08 Quentin Anthony So we released a blog at Zyphra on porting flash attention to over to my 300 X, just to sort of play out, can we move our own stack over 4 times in the scene
- ▶ 5:26 unnamed speaker What's the state of, um, AMD savior? 4 times in the scene
- ▶ 6:39 Quentin Anthony The hardware of MI three to next has these properties and I want my algorithm to pull these properties out 2 times in the scene
- ▶ 10:29 unnamed speaker What are like really the things that in your mind are like, okay, this is like a huge advantage versus like, this is like part of the offering, but like I can replicate that somewhat easily on AMD or whatever else. 4 times in the scene
- ▶ 57:28 Quentin Anthony Um, I would love to get more AMD sort of work and like alternative ecosystem stuff. 3 times in the scene
⚡ Open Model Pretraining Masterclass — Elie Bakouch, HuggingFace SmolLM 3, FineWeb, FinePDF
- ▶ 1:01:49 Shawn Wang Like, you know, if you've got some AMD, you want to use that.
Why RL Won — Kyle Corbitt, OpenPipe (acq. CoreWeave)
- ▶ 49:00 Shawn Wang And then even AMD deal.
⚡️Accelerators @ 3x NVIDIA H200 perf, Made in the USA - Thomas Sohmers + Mitesh Agrawal, Positron AI
- ▶ 19:48 Thomas Sohmers And, you know, be it a lot of the, the companies that you mentioned and, and, and others, I mean, even big guys like AMD have more or less 4 times in the scene
- ▶ 35:07 Mitesh Agrawal And if our architecture is suitable for certain industries much more than let's say NVIDIAs or AMDs or any other silicon architectures, I think they would use us. 2 times in the scene
- ▶ 40:04 Thomas Sohmers Uh, um, you know, from, it's, it's not just from, as Matesh was saying, that from a customer's perspective, if they're deploying hardware, they're going to be choosing between be it NVIDIA, AMD, Positron, or anyone else.
The Shape of Compute (Chris Lattner of Modular)
- ▶ 5:21 Chris Lattner Uh, we're about to launch our AMD and my 303 25 support.
- ▶ 5:21 Chris Lattner Uh, we're about to launch our AMD and my 303 25 support.
- ▶ 11:34 Chris Lattner So it's not as good as something like VLLM because it's missing some features, and it only supports NVIDIA and AMD hardware, for example. 4 times in the scene
- ▶ 24:38 Chris Lattner We support Google TPUs and Infantria and AMD and NVIDIA, obviously, and CPUs and this and that and the other thing.
AI Engineering for Art - with comfyanonymous
- ▶ 33:52 Shawn Wang Like, is it like 70% NVIDIA and like 30% AMD 6 times in the scene
- ▶ 34:43 comfyanonymous (Comfy) So, until AMD actually, like, uh, ports their, like, raw cam to, to Windows properly, and then there's actually PyTorch, I think they're, they're doing that, they're in the process of doing that, but, uh, until they get it, they get a… 2 times in the scene
llm.c's Origin and the Future of LLM Compilers - Andrej Karpathy at CUDA MODE
- ▶ 20:53 Andrej Karpathy A and B fork is very active, as far as I understand, and quite good.
The Winds of AI Winter (Q2 Four Wars of the AI Stack Recap)
- ▶ 20:44 unnamed speaker You know, AMD, like Maddox, like, um, Etched, you know, which, which caused a lot of noise with their Sohu chip as well.
Supervise the Process of AI Research — with Jungwon Byun and Andreas Stuhlmüller of Elicit
- ▶ 52:39 unnamed speaker I don't know if that's like the best name to, to think about it, but I was using one of these chat which are documents features, and I put the AMD and my 300 specs on the, um, new, you know, Blackwell chips from NVIDIA, and I was asking… 2 times in the scene
- ▶ 52:39 unnamed speaker I don't know if that's like the best name to, to think about it, but I was using one of these chat which are documents features, and I put the AMD and my 300 specs on the, um, new, you know, Blackwell chips from NVIDIA, and I was asking…
Open Source AI is AI we can Trust — with Soumith Chintala of Meta AI
- ▶ 10:42 unnamed speaker Um, I think George is trying to make, especially AMD stack be better than, than Rockham. 3 times in the scene
- ▶ 57:49 Soumith Chintala Is they, each large company has a sufficient enough set of verticalized workloads, uh, that have a pattern to them that, say, a more generic accelerator like an NVIDIA or an AMD GPU does not exploit.
The AI-First Graphics Editor - with Suhail Doshi of Playground AI
- ▶ 6:18 Suhail Doshi And there's like three CPU manufacturers, two of which sell that, you know, you know, big ones, you know, AMD, Intel, and then of course, like Apple made the M one.
The State of Silicon and the GPU Poors - with Dylan Patel of SemiAnalysis
- ▶ 41:13 Dylan Patel Um, AMD. 8 times in the scene
- ▶ 41:16 Dylan Patel MI 300. 2 times in the scene
- ▶ 48:06 Dylan Patel Like, I think, I think, uh, I think OpenAI would be stupid not to do that with NVIDIA, but also with AMD and, but also, hey, like how much time, and, and Microsoft's internal silicon, but it's like, how much time do I actually have, right?
- ▶ 59:21 Dylan Patel AMD's MI 300 piece, it talks a lot about, the one that we did on that, are very good.
- ▶ 59:21 Dylan Patel AMD's MI 300 piece, it talks a lot about, the one that we did on that, are very good.
FlashAttention-2: Making Transformers 800% faster AND exact
- ▶ 33:28 unnamed speaker So like could you use this and then run it like a AMD GPU? 2 times in the scene
Ep 18: Petaflops to the People — with George Hotz of tinycorp
- ▶ 22:34 George Hotz And then AMD. 5 times in the scene
- ▶ 32:37 George Hotz Unless AMD, you know, upsets me again, and then we're, you know, we're back to, we're back to other colors.