AI/ML
Artificial Intelligence & Machine Learning
Build
Explore/Develop/Train
AI Training/Inference
- Cerebro: 6 Alveo FPGAs
- Genisys: 6 V100 GPUs
- Synergy: 2 GPU 2 FPGA
- HW emulation, acceleration, validation
- FPGA Prototyping and SW bring-up
AI Training
- 32 Ascend 910
- 16 Kunpeng 920
- 8 PFLOPS@FP16
- CANN computing architecture
- SW stack allowing 32 simultaneous users
Optimize
Prototype & Benchmark
AI-DRIVEN OPTIMIZER
Optimize Deep Neural Networks by making them faster, smaller and energy-efficient from cloud to edge computing
An AutoML studio to create faster, smaller, more power efficient deep learning models for various hardware backends.
Deploy
Landing your solution
Edge PLATFORMs and IPs
32- and 64-bit open-source RISC-V cores
- Barrel RISC-V Processor For AI Acceleration
- RISC-V Processor Design and Prototyping
Atlas 200 DK
AI Developer Kit
- AI @ edge
- Ascend 310
- 22 TOPS INT8
16 TOPS INT8
8 TOPS INT8
News
Events
may 2022
