Edge AI
Effortlessly Train, Optimize and Deploy ML models on edge devices
Build
AI Training
CLOUD
Features
- 4 nodes delivering 8 PFLOPS
- FP16 2.56 PFLOPS/5.6KW/4U
- 8 100G RoCE v2 high-speed ports
Applications
- Model training
- HPC
- Smart city Smart healthcare
- Astronomical Exploration
- Oil exploration
Features
- Tesla V100 Nodes
- Alveo U250 Nodes
- 8 nodes
Applications
- Model training
- HPC
- ASICs Prototyping
- Custom neural network accelerators
Optimize
AI Optimization
AI-DRIVEN OPTIMIZATION
AI Optimization
Optimize Deep Neural Networks by making them faster, smaller and energy-efficient from the cloud to the edge
Supported ML Frameworks:
- TensorFlow
- PyTorch
- KERAS
- Caffe
- ONNX
- TVM
Deploy
AI Inference
EDGE
Features
- 502 INT8 TOPs
- 204MB on-chip SRAM
- 75W TDP 40W typical
- 8 TOPs/W
Applications
- Vision
- NLP
- Finance
- HPC
Atlas 200 DK
- Custom AI accelerator
Applications
- Low Power Edge AI
- Mobile biometrics
- ADAS
- Industrial automation
- Vision
Partnerships with AI Hardware and Software Leaders
News
Events
october 2023
Atlas 800 AI
Student Competition
$50,000 in Total Prize Money
Submission Deadline: JAN 15
Submission Deadline: JAN 15
NEW!
AI Optimization icon made by Eucalyp from www.flaticon.com