SC23 Daily Roundup: Day 1

Written by:

Monday, November 13th

We’re excited to be exhibiting at Supercomputing for the second year in a row, with Groqsters from all over the world, from California to the UK, attending the major industry conference this week in Denver, Colorado. This week, our team is coming together to present during SC23, conduct partner tutorials with our customers, host 1:1 meetings in our VIP lounge, and share demos of our Language Processing Unit™ Inference Engine live with attendees at our booth. 

This year, we’re located at booth 1681 and when the exhibition hall opened at 7pm, the team greeted the first wave of folks to come through. Groqsters talked to hundreds of leaders, engineers, researchers, and students and shared our GroqLabs demo running Llama-2 70B at over 300 tokens per second per user. 

We continued to hear “WOW!” from folks seeing the demo for the first time, our friend Dr. Ian Cutress of TechTech Potato came to meet the team, and many stopped to see and ask questions about the GroqCard™ accelerator and GroqNode™ server. 

This wasn’t the beginning of our conference activities though. On Sunday, Sanjif Shanmugavelu, one of our software engineers, represented Groq at Argonne National Laboraory’s tutorial, Programming Novel AI Accelerators for Scientific Computing. 

During this session, Sanjif educated ~50 potential Groq system users about our capabilities now available to researchers through Argonne. Presenting alongside SambaNova, Graphcore, Cerebras, and Intel, Groq was the only accelerator to showcase a live demo, generating quite the buzz and interest post-talk. As we move closer and closer to our December 6 & 7 Groq AI Workshop, more users are hearing about the opportunity to apply to use Groq through the ALCF AI Testbed.

To see more from Groq tomorow, don’t miss the following demos at our booth.

Demos at Booth 1681 | Tuesday, November 14, 2023

10:00–10:30am MST

  • Record Setting LLM Performance – 300+ Tokens/s per User on a LPU™ Inference Engine 
  • Graham Steele & Mark Heaps

1:00–1:30pm MST

  • Ultra-low Latency Inference with GroqChip™ Accelerator + FPGA
  • Tobias Becker 

3:00–3:45pm MST

  • Accelerating AI for Science with Groq at ALCF AI Testbed
  • Murali Emani, ALCF Computer Scientist, Argonne National Laboratory

Never miss a Groq update! Sign up below for our latest news.

The latest Groq news. Delivered to your inbox.

The latest Groq Developer news. Delivered to your inbox.