Groq

Join us next week in New York! Mark Heaps and Hatice Ozen will be presenting and Paul Piezzo will be at our table to show you some amazing demos and answer your questions.

Our engineering team is cooking - latency, throughput and quality all keep improving across different models. 🚀 🙏 artificialanalysis.ai for the independent benchmarking!

Tune in remotely at 8:00 PT on July 10 for a TPC seminar by Valentin Reis. Learn how Groq co-designed a compilation-based software stack and a class of accelerators called LPUs, with high utilization and low end-to-end system latency. He’ll review the challenges of breaking models apart over networks of LPUs, and outline how this hardware/software system architecture keeps enabling breakthrough LLM inference latency at all model sizes. https://lnkd.in/gdQBuS7U

Ross claims that by next year, over half of the globe’s inference computing will be running on their chips. He’s going to be diving into why #AI inference is critical to enterprise technology, why efficiency is today’s urgent need — and why Groq is at the forefront of that revolution. He’ll be offering one of those mind-blowing live demos to highlight Groq’s capabilities, and talk about what’s next on the roadmap.

Tune in to the upcoming TPC seminar by Valentin Reis to learn how Groq co-designed a compilation-based software stack and a class of accelerators called Language Processing Units, with high utilization and low end-to-end system latency. He’ll review the challenges of breaking models apart over networks of LPUs, and outline how this hardware/software system architecture keeps enabling breakthrough LLM inference latency at all model sizes. This is a remote event - join live at 8:00 PT at https://hubs.la/Q02BD6PH0

We recently started offering the speech-to-text model Whisper V3 Large on GroqCloud. Now you can experience the speed of it yourself in GroqChat. "digital assistance everywhere through AI is going to change interfaces completely. Voice is the logical first step of that progression and now people can build that on Groq." - Mark Heaps, Chief Tech Evangelist and VP of Brand. Go to Groq.com to try it yourself.

If you missed Sunny Madra, GM of GroqCloud, speaking about breaking AI's 1 gigahertz barrier at AI Engineer World's Fair, you can catch a recording of his main stage presentation here: https://hubs.la/Q02DQ0XL0.

WDYT?

Last week, we at Berkeley SkyDeck, together with Cal Hacks, organized the largest in-person hackathon in the US! We received 2700+ applications, had 1200 hackers, 40 organizers, 140 judges (including myself), 80 mentors, 11 sponsors, and 60 volunteers. We were honored to have Andrej Karpathy (Founding member of OpenAI / Former Senior Director of AI at Tesla) as a speaker. This event was made possible thanks to sponsors like Groq, Intel Corporation, Mayfield Fund, Lightspeed, Amazon Web Services (AWS), Hume AI, SingleStore, You.com, Reach Capital, Berkeley Frontier Fund Start rocket, and Academic Innovation Catalyst, as well as partners like Mistral AI, OpenAI, MindsDB, Agora, LMNT, Feedback Intelligence, Luxius, Floz, and Biskout. Here are the winning teams: 👑 Grand prize by SkyDeck Fund: Dispatch AI 💻Intel : Dispatch AI, Batteries by LLM, Accel 🤖Groq: Scam Scanner ☁️AWS: SafeGuard, GPTEthics & DPAncestry, DisasterAid.ai 👍AI for good by @AIC: Dispatch AI, ASL Bridgify 🌍Climate Tech Track: Greenwise 🫀Hume: Lock in, Bloom Buddy, ScamScanner Thank you Aaron Melgar Jean Farmer Rohit Talluri ☁️ Kevin Lu Matt Sonsini Paul Work Gabrielle Amoranto Omar Toral Rahul Unnikrishnan Nair Benjamin Odom Alex Sin Eduardo Alvareze Alex P. Sarit Schube Rick Lamers Jose M. Zach Greathouse Jeremy Hadfield Dylan Chow Nicolas Schuhl Sophia Yang, Ph.D. Yansong Pang

🙏 Annie Liao 🇦🇺 We are looking forward to many more collaborations with you and Build Club 🚀