DeepSeek’s R1 70B combines the powerful reasoning ability of the full R1 model with the size and speed of Llama 70B. R1 70B outperforms GPT-4o and o1-mini across a range of general and reasoning benchmarks, making it the most capable Llama 70B variant by far.
@cerebras
-
DeepSeek R1 70B Now Available on Cerebras Infrastructure
By
–
DeepSeek R1 70B is now on Cerebras!
– Instant reasoning at 1,500 tokens/s – 57x faster than GPUs
– Higher model accuracy than GPT-4o and o1-mini
– Runs 100% on Cerebras US data centers https://
inference.cerebras.ai -
CerebrasCoder 60-Second Hackathon: Win $250 Prizes
By
–
2025 is the year of 60-second hackathons! This is your chance to win one of multiple $250 prizes…in less than 1 minute. We've had 100,000+ projects built on CerebrasCoder and are excited to announce the CerebrasCoder hackathon. CerebrasCoder is an open-source app that
-
Breakthrough AI Conversation with Jean Philippe Fricker
By
–
The thrill, the terror, and a breakthrough. Listen here:
Apple Podcasts: https://
bit.ly/4aobCFV
Spotify: https://
bit.ly/3CdZuKU Thank you @marcelsalathe and the @EPFL_en AI Center for this fascinating conversation with Jean Philippe Fricker. -
Cerebras and Mayo Clinic Launch Genomics Foundation Model
By
–
Cerebras and @MayoClinic are proud to announce a new state-of-the-art foundation model for genomics.
— Cerebras (@cerebras) 14 janvier 2025
The breakthrough was made possible by combining Mayo Clinic's extensive patient data with the Cerebras AI platform, training on over a trillion tokens to create complex genomic… pic.twitter.com/4LEUhimiLQCerebras and @MayoClinic are proud to announce a new state-of-the-art foundation model for genomics. The breakthrough was made possible by combining Mayo Clinic's extensive patient data with the Cerebras AI platform, training on over a trillion tokens to create complex genomic
-
Mayo Genomic Model Achieves Unprecedented Disease Prediction Accuracy
By
–
The model achieves unprecedented accuracy in disease prediction: 87% for Rheumatoid Arthritis
96% for cancer predisposition
83% for cardiovascular conditions. The Mayo Genomic Foundation Model is a significant step forward in advancing precision medicine. -
WSE-3 Wafer Scale Engine Achieves 93% Silicon Utilization
By
–
Despite being the world’s largest computer chip, the WSE-3 achieves 93% silicon utilization—higher than today’s leading GPU. A tiny core, a self-healing fabric, and a clean software abstraction – that’s how you do wafer scale.
-
WSE Defect Tolerance: 164x Better Silicon Efficiency
By
–
One way to measure defect tolerance is to look at the amount of die area. Assuming 0.001 defects per mm^2, GPUs will see 59 defects disabling 361mm^2 of silicon. On the WSE, it sees 46 defects and loses just 2.2mm^2. That's a 164x reduction in dead silicon area!
-
H100 vs WSE: Wafer-Scale Architecture Comparison at 5nm
By
–
Let’s compare at the wafer level. At 5nm, here’s what an H100 sized chip and WSE look like side by side. A 300mm wafer yields ~72 GPUs or one Wafer Scale Engine.
-
Fault-Tolerant Network Fabric with Redundant Communication Pathways
By
–
Our network fabric is also fault tolerant. When a defect is detected, the chip automatically routes around it using redundant communication pathways. The software layer presents the chip as a 2D grid even when there are defects.