A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence · HackerPapers discussions

A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence — mentioned by westurner

Wed, 08 Oct 2025 18:33:08 GMT

> L3 cache is not built for mass throughput in the same way that DRAM is, and so it has roughly identical mass throughput despite its much closer distance to the computation. "The von Neumann bottleneck is impeding AI computing?" (2025) https://news.ycombinator.com/item?id=45398473 : > How does Cerebras WSE-3 with 44GB of 'L2' on-chip SRAM compare to Google's TPUs, Tesla's TPUs, NorthPole, Groq LPU, Tenstorrent's, and AMD's NPU designs? From https://news.ycombinator.com/item?id=42875728 : > WSE-3: 21 PB/S From https://hackernoon.com/nvidias-mega-machine-crushes-all-of-2... : > At Computex…

Paper and discussion history. A mention is not an endorsement.

A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence — mentioned by sailingparrot

Wed, 01 Oct 2025 15:42:10 GMT

> What about for inference? Same techniques apply. > but did not dispute that they store the entire model on SRAM. No idea what they did or did not do for that specific test (which was about delivering 1800 tokens/sec though, not simply running qwen-3) since they didn't provide any detail. I don't think there is any point storing everything in SRAM, even if you do happen to have 100M$ worth of chips lying around in a test cluster at the office, since WSE-3 is designed from the ground up for data parallelism (see [1] section 3.2) and inference is sequential both within a single token…

Paper and discussion history. A mention is not an endorsement.

A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence — mentioned by westurner

Thu, 21 Aug 2025 22:51:58 GMT

TPUs are more efficient at LLMs because they do more TOPS/KWh. "OpenTPU: Open-Source Reimplementation of Google Tensor Processing Unit (TPU)" (2025) https://news.ycombinator.com/item?id=44111452 "A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence" (2025) https://arxiv.org/html/2503.11698v1 From https://news.ycombinator.com/item?id=44648575 : > "Next-generation datacenters consume zero water for cooling" (2024) https://news.ycombinator.com/item?id=42376406 >> this design will avoid the need for more than 125 million liters…

Paper and discussion history. A mention is not an endorsement.

A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence — mentioned by skryl

Sat, 31 May 2025 12:57:01 GMT

Performance per watt is better than h100 and b200, performance per watt per $ is worse than B200, and it does fp8 just fine https://arxiv.org/pdf/2503.11698

Paper and discussion history. A mention is not an endorsement.

A Comparison of the Cerebras Wafer-Scale Integration Technology with Nvidia GPU-based Systems for Artificial Intelligence — HN submission

Mon, 31 Mar 2025 17:48:00 GMT

Cerebras Wafer-Scale Integration vs. Nvidia GPU-Based Systems for AI

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.