Community meetings

Next Community Meeting July 12, 2022

This month we have a special developer community meeting prepared for you! One of our Senior Deep Learning Engineers, Sia Rezaei, will discuss our latest capability – easily training 1.3 – 20 billion parameter GPT models on a single CS-2 machine. This is record-setting for large language model trainings on a single device.

For more information, please see the following articles:

    • Our announcement blog, which explains how easy it is to train Multi-Billion Parameter Models on a Single Cerebras System.
    • Tom’s Hardware, “Cerebras Slays GPUs, Breaks Record for Largest AI Models Trained on a Single Device.”
    • VentureBeat, “By enabling a single CS-2 to train these models, Cerebras reduces the system-engineering time necessary to run large NLP models from months to minutes.”

We hope that you can join us for this exciting presentation and discussion!

Location: via Zoom.


Register here

How you can contribute

The Cerebras Develop Community is not a spectator sport (unless you want it to be). We urge you to speak up and get involved.

From low effort to more effort …

  • Attend meetings!
  • Ask questions (we do chat Q&A during presentations then a live session at the end)
  • Participate in our surveys
  • Answer questions from your peers
  • Share work you have done on our systems
  • Lead a training seminar

Calling all SDK Users!

We are seeking volunteers to help with an SDK user study. If you are an active developer writing code with the Cerebras SDK, we are interested in hearing from you.

To participate, please email us at developer@cerebras.net with the subject “User Study Volunteer” to indicate your interest, and we’ll schedule a 30-minute session with you.