See the future of AI at #ISCWest2025! Axelera AI’s Metis™ AI Platform delivers: 795 FPS on YOLOv5s LLM acceleration w/ Arduino Live demos & innovations April 2-4 | Venetian Expo, Las Vegas Booth 32087 Book a meeting: https://
eu1.hubs.ly/H0hFNp80 @ISCEvents
COMPUTING
-
Axelera AI Metis Platform Showcases 795 FPS YOLOv5s Performance
By
–
-
COBOL Unlearning and AI Model Training Implications
By
–
“How could unlearning COBOL lead to this?” I’m crying
-
Microprocessors and AI: The Global Battle Nvidia AMD Huawei
By
–
New on the Blog: Microprocessors and Artificial Intelligence: The Global Challenge between Nvidia, AMD and Huawei Stay ahead in the ever-evolving world of Digital Transformation. This article explores key insights into emerging technologies and their impact on today's business
-
Dynamo: Open-Source High-Throughput AI Inference Library
By
–
10. Dynamo: A high-throughput, low-latency open-source inference library to deploy AI reasoning models in large-scale distributed environmentspic.twitter.com/shSKQmUoca
— Rowan Cheung (@rowancheung) 20 mars 202510. Dynamo: A high-throughput, low-latency open-source inference library to deploy AI reasoning models in large-scale distributed environments
-
DGX Station: High-Performance AI Computer for Researchers
By
–
7. DGX Station: A high-performance AI computer for data scientists and researchers
— Rowan Cheung (@rowancheung) 20 mars 2025
20 petaflops of performance with 72 CPU cores, a chip-to-chip interface, high-bandwidth memory, and PCIe slots for GeForcepic.twitter.com/GbrREYVxGP7. DGX Station: A high-performance AI computer for data scientists and researchers 20 petaflops of performance with 72 CPU cores, a chip-to-chip interface, high-bandwidth memory, and PCIe slots for GeForce
-
DGX Spark: World’s Smallest AI Computer with Grace Blackwell
By
–
6. DGX Spark: The world's smallest AI computer
— Rowan Cheung (@rowancheung) 20 mars 2025
1000 TOPS of AI performance with 128GB RAM and Grace Blackwell architecturepic.twitter.com/UNWR1MZ4Fi6. DGX Spark: The world's smallest AI computer 1000 TOPS of AI performance with 128GB RAM and Grace Blackwell architecture
-
NVIDIA GPU Roadmap: Blackwell, Vera Rubin, Feynman Through 2028
By
–
4. Future GPU roadmap: After Blackwell Ultra, Vera Rubin and its ultra version in 2026-27, followed by Feynman in 2028 — each promising exponential performance gains to meet surging computational demands!pic.twitter.com/IamlzPjP2z
— Rowan Cheung (@rowancheung) 20 mars 20254. Future GPU roadmap: After Blackwell Ultra, Vera Rubin and its ultra version in 2026-27, followed by Feynman in 2028 — each promising exponential performance gains to meet surging computational demands!
-
Blackwell Ultra: Next-Gen GPU with 1.5x Computational Power
By
–
3. Blackwell Ultra: The next generation of Blackwell with 1.5x computational power — coming in the second half of 2025pic.twitter.com/bGWkgzHS0p
— Rowan Cheung (@rowancheung) 20 mars 20253. Blackwell Ultra: The next generation of Blackwell with 1.5x computational power — coming in the second half of 2025
-
NVIDIA GTC 2025: Blue Robot and Physics Engine Breakthrough
By
–
It's been 4 days of NVIDIA GTC 2025, and the announcements have been incredible.
— Rowan Cheung (@rowancheung) 20 mars 2025
The 10 most important reveals so far:
1. Blue: A Star Wars-inspired robot powered by a new physics engine with real-time intelligence and movementpic.twitter.com/3D63moWUdaIt's been 4 days of NVIDIA GTC 2025, and the announcements have been incredible. The 10 most important reveals so far: 1. Blue: A Star Wars-inspired robot powered by a new physics engine with real-time intelligence and movement
-
GridMix Enhances Implicit Neural Representations for PDE Solving
By
–
This ICLR 2025 Oral paper from Tsinghua University presents GridMix, which enhances implicit neural representations (INRs) for solving PDEs.