Linear Scaling Made Possible with Weight StreamingIn a single keystroke, Cerebras Systems can scale large language models from a single CS-2 system to 192 CS-2s in a Cerebras Wafer-Scale…Sep 15, 2022Sep 15, 2022
Cerebras Architecture Deep Dive: First Look Inside the HW/SW Co-Design for Deep LearningOur ML-optimized architecture enables the largest models to run on a single device. With data parallel-only scale out and native…Aug 30, 20221Aug 30, 20221
Context is Everything: Why Maximum Sequence Length Matters for AIGPU-Impossible™ sequence lengths on Cerebras systems may enable breakthroughs in Natural Language Understanding, drug discovery and…Aug 17, 20222Aug 17, 20222
A Big Chip for Big Science: Watching the COVID-19 Virus in ActionIt’s hard to imagine a better example of “AI for good” than figuring out how the virus that causes COVID-19 works. If we know how it works…Jul 29, 2022Jul 29, 2022
When Time is Money: Accelerating NLP Model Training at a Leading Financial Institution with…Our Cerebras CS-2 system delivered the compute performance of more than 120 AI-optimized GPUs, making it possible to train large BERT…Jul 27, 2022Jul 27, 2022
If You’re Doing Pharma and Life Sciences AI Research Without a Cerebras System, You’re Doing it…Cerebras AI accelerator systems make it possible to train complex models using huge datasets at the speed of science.Jul 26, 2022Jul 26, 2022
Cerebras Sets Record for Largest Artificial Intelligence Models Ever Trained on Single DeviceOur customers can easily train and reconfigure GPT-3 and GPT-J language models with up to 20 billion parameters on a single CS-2 systemJul 25, 2022Jul 25, 2022
Training Multi-Billion-Parameter Models on a Single Cerebras System is EasyChanging model size is trivial on Cerebras, rather than a major science project using conventional graphics accelerators.Jul 21, 2022Jul 21, 2022
Cerebras Makes It Easy to Harness the Predictive Power of GPT-JA look at why this open-source language model is so popular, how it works and how simple it is to train on a single Cerebras system.Jul 20, 2022Jul 20, 2022
Increasing Model Throughput with Variable Tensor Shape ComputationsWe write machine learning algorithms to fit the data, not pad the data to suit hardware limitations.Jun 6, 2022Jun 6, 2022
