Google TPU, every mention
11 scenes, the whole family · ← back to Google TPU
tap a year for its mentions
every year anyone Alessio Fanelli 9Dylan Patel 8George Hotz 2Sean Lie 1Chris Lattner 1
Verbatim, from the transcripts: the passages where Google TPU comes up
The Inference Frontier: from 100 to 10,000 tokens per second — Sean Lie, Cerebras CTO
⚡️Raising $1.1b to build the fastest LLM Chips on Earth — Andrew Feldman, Cerebras
- ▶ 13:25 unnamed speaker Yeah, I think that the numbers I had was something internally for Google with the TPU workload.
The Shape of Compute (Chris Lattner of Modular)
- ▶ 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.
The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka
- ▶ 48:30 unnamed speaker I think you addressed this in your post, but the temptation would have been just to run everything on TPUs, which is the stack that you already know very well, that works very well. 6 times in the scene
The State of Silicon and the GPU Poors - with Dylan Patel of SemiAnalysis
- ▶ 5:45 Dylan Patel Me, myself, and my colleagues is like, we go from software all the way through to like low-level manufacturing, and it's like, who, you know, oh, Google's actually ramping up TPU production massively, right?
- ▶ 10:54 Alessio Fanelli That makes sense, and the, the chart that Sama mentioned, uh, is about, yeah, Google TPU B-Fives completely overtaking OpenAI by, like, orders of magnitude. 5 times in the scene
- ▶ 10:54 Alessio Fanelli That makes sense, and the, the chart that Sama mentioned, uh, is about, yeah, Google TPU B-Fives completely overtaking OpenAI by, like, orders of magnitude. 4 times in the scene
- ▶ 11:49 Dylan Patel Like some of the numbers they've, they've shown, some of the, you know, code they've shown for TPUv-V-E, which is not the TPUv-V-V that I was referring to, which is, uh, in, in the sort of the post, the GPU poor post I was referring to,… 3 times in the scene
- ▶ 49:34 Dylan Patel Difficult to see anyone truly pivoting to anything besides a GPU and a TPU, especially if you have, if you need that scale, right? 2 times in the scene
- ▶ 59:09 Dylan Patel I think that one's really, uh, uh, critical because it explains Google's infrastructure, uh, quite a bit from networking through chips, through all that sort of history of the TPU a little bit, and all this sort of stuff. 2 times in the scene
Ep 18: Petaflops to the People — with George Hotz of tinycorp
- ▶ 1:14:33 George Hotz Google TPUs were actually the ultimate. 2 times in the scene