StarZero All shows Try the podcast clipper →

Cerebras Systems

also referred to as: cerebras

22 statements across 3 episodes · 14 bullish · 0 bearish · 3 people on the record · first statement Feb 20, 2020 by Michael James · said 69 times in 10 episodes since 2019 · across every show →

Mentions by year

brought up most by Matt Turck (42), Andrew Feldman (12), Sanjit Biswas (4), Dylan Patel (4), Michael James (3), Tim Dettmers (1), Erik Bernhardsson (1), Dillon Erb (1)

tap a year for its mentions
0030360520192020202120222023202420252026episodesmentions
03520192020202120222023202420252026episodes it came up in
007.52.515520192020202120222023202420252026episodesmentions per episode
2026 59 mentions in 5 episodes 12 per episode
2024 6 mentions in 3 episodes 2 per episode
2020 3 mentions in 1 episode
2019 1 mention in 1 episode

every mention, scene by scene, with the transcript →

Everything said about Cerebras Systems, oldest first

Feb 20, 2020 bullish
Assertion Supported
Cerebras's chip provides 10,000x the memory bandwidth of standard GPUs
“So, having solved this, we were left with a machine that was pretty impressive on most of the metrics, being larger, more cores, 10,000 times the memory bandwidth and 33,000 times the fabric bandwidth.”
Michael James Feb 20, 2020 ▶ 13:10 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020
Disclosure
Cerebras designed a chip as large as possible with interleaved memory
“We decided to build a machine that was as large as we could, that had memory interleaved directly with the computational substrate.”
Michael James Feb 20, 2020 ▶ 6:44 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 positive
Assertion Supported
Cerebras's first customer was the U.S. National Labs
“It's pretty unusual in Silicon Valley to have your first customer be the U.S. National Labs”
Michael James Feb 20, 2020 ▶ 14:46 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 positive
Assertion Contradicted
Cerebras built the first supercomputer designed end-to-end for AI
“And it's also the first supercomputer that has been designed from beginning to end for artificial intelligence and deep learning.”
Michael James Feb 20, 2020 ▶ 0:33 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 positive
Assertion Supported
Cerebras routes around manufacturing defects across its 400,000 cores
“By having 400,000 independent cores, it's perfectly fine if we have a few hundred defects on there, and it's only our job to route the information flows around those defects.”
Michael James Feb 20, 2020 ▶ 8:53 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 neutral
Assertion Supported
Cerebras' Wafer Scale Engine is fabricated at TSMC
“It is fabricated at Taiwan Semiconductor, which is one of the world's large major fab houses.”
Michael James Feb 20, 2020 ▶ 18:23 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 positive
Opinion
Programming Cerebras's engine is simpler than programming conventional computers
“So again, the simpler, ah, I think simpler than programming a conventional computer where you have to specify millions of instructions.”
Michael James Feb 20, 2020 ▶ 13:29 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 neutral
Assertion Supported
No motherboard had ever housed a processor as large as Cerebras's
“No one before had ever put a processor that large in a motherboard, and so we had to also make a system to house this thing.”
Michael James Feb 20, 2020 ▶ 9:19 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020
Assertion Not checkable as stated
Wafer-scale chips were previously presumed impossible to build
“And when we showed this the world got pretty excited, so there was a large press splash, and in part that's because this had been presumed to be impossible to achieve.”
Michael James Feb 20, 2020 ▶ 8:19 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 positive
Assertion Not checkable as stated
Michael James: Cerebras succeeded at reticle stitching on the first attempt
“Reticle stitching to get above the optical limit of how you manufacture these things that worked the first time we tried it.”
Michael James Feb 20, 2020 ▶ 10:50 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020
What-if
Traditional power delivery would carry 40,000 amps and melt Cerebras's processor
“We would have 40,000 amps, and that would actually melt the part if we chose to do it that way, so we had to bring some of the geometric thinking beyond how we think geometry in the models into this three D structure.”
Michael James Feb 20, 2020 ▶ 10:02 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020
Assertion Supported
Cerebras maps neural network layers across 400,000 chip cores
“We have, ah, an optimization solver that places these over the 400,000 cores.”
Michael James Feb 20, 2020 ▶ 14:21 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Feb 20, 2020 positive
Assertion Supported
Lawrence Livermore integrated Cerebras into its Lassen supercomputer
“Lawrence Livermore has integrated into their Lassen supercomputer.”
Michael James Feb 20, 2020 ▶ 15:08 Designing an AI Supercomputer // Michael James, Cerebras (FirstMark's Data Driven NYC)
Jul 23, 2026 positive
Disclosure
Feldman: Cerebras signed a 760-megawatt multi-year compute deal with OpenAI
“The deal is 760 megawatts, 250 megawatts in 26 on a multi-year lease. An additional 250 megawatts in 27, on a multi-year lease, and an additional in 28, a multi-year lease.”
Andrew Feldman Jul 23, 2026 ▶ 56:26 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 neutral
Disclosure
Feldman: Cerebras burned $8M monthly for 18 months before building successful chips
“And we had a, an 18 month period where we were spending eight million a month and we couldn't build them.”
Andrew Feldman Jul 23, 2026 ▶ 34:19 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 bullish
Assertion Contradicted
Feldman: Cerebras sales were 10x higher than Groq's at acquisition
“And we were the fastest at it, and the largest, and, you know, our sales were more than 10 times the Grox, and they paid twenty billion dollars for the number two collector.”
Andrew Feldman Jul 23, 2026 ▶ 9:01 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 positive
Assertion Supported
Feldman: Cerebras built a 46,000 square millimeter wafer-scale chip
“The biggest chip that had ever been built before us was 800 square millimeters. 840 to be exact. And this is 46,000.”
Andrew Feldman Jul 23, 2026 ▶ 33:41 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 positive
Assertion Partly supported
Feldman: Cerebras completed the largest semiconductor IPO in history on May 14th
“And so when we rang the bell and we went public on May 14th this year and the largest semiconductor IPO in history and we did something unusual.”
Andrew Feldman Jul 23, 2026 ▶ 37:56 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 bullish
Assertion Supported
Feldman: Cerebras moves weights to compute ~2,500x faster than standard GPUs
“And so the speed of moving waits to compute is about two and a half thousand times faster here than on a Wilben GP.”
Andrew Feldman Jul 23, 2026 ▶ 44:43 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 neutral
Disclosure
Feldman: Cerebras serves second-tier AI labs for model training
“We do RL and we do traditional training too. Not for the largest models, for the largest lab, but for the next tier.”
Andrew Feldman Jul 23, 2026 ▶ 45:43 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 23, 2026 bullish
Disclosure
Feldman: Cerebras signed an OpenAI compute deal worth over $20 billion
“Remember, we did a huge deal. This is probably the largest deals in Silicon Valley history north of twenty billion dollars.”
Andrew Feldman Jul 23, 2026 ▶ 9:49 Cerebras CEO: Why GPUs Can't Do Fast Inference
Jul 29, 2026 positive
Assertion Supported
Cerebras cloud achieves 10x inference speedup over fast GPUs on Gemma
“Say, if you run Gemma four, On your GP, you might get like a hundred tokens per second if you have a fast card. If you run it in their cloud, you get anywhere from 800 to 1500 tokens per second. So call it 10 X faster.”
Sanjit Biswas Jul 29, 2026 ▶ 44:58 The Biggest AI Deployment Nobody Talks About | Samsara CEO Sanjit Biswas
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 400 conversations transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more