Cover Image for Cerebras Tech Talk Series: Deep Dive into Deepseek! 🖥️
Cover Image for Cerebras Tech Talk Series: Deep Dive into Deepseek! 🖥️
Avatar for Cerebras
Presented by
Cerebras
Cerebras makes the world's biggest chip, powering fast inference up to 15x faster than GPUs.
393 Went
Sneha Khanvilkar
invites you to join

Cerebras Tech Talk Series: Deep Dive into Deepseek! 🖥️

Virtual
Registration
Past Event
Welcome! To join the event, please register below.
About Event

​Join us for the first event in the Cerebras Tech Talk Series

​Join Daria Soboleva, Head Research Scientist at Cerebras, who will lead a talk introducing MoE (Mixture of Experts), how Deepseek's models leverage MoE architecture, and what this means for the future of AI models.

​Where: [Virtual] Cerebras Discord

  1. ​Join the server: Cerebras.ai/Discord

  2. ​Join the event: Join Discord Server

​The Cerebras Tech Talks are a technical series presented by Cerebras. Every other week, we invite an AI researcher or engineer to present on the latest, most relevant AI topics.

​Cerebras is building the largest and fastest AI processor in the world. If you are interested in building with the fastest inference (serving Llama 3.3 70B at 2100 tok/s)

​Email sarah.chieng@cerebras.net to get started.

​The event isn’t limited to computer programmers. Anyone with an interest in AI/ML is encouraged to participate!

Avatar for Cerebras
Presented by
Cerebras
Cerebras makes the world's biggest chip, powering fast inference up to 15x faster than GPUs.
393 Went