Statements (60)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:microprocessor
|
| gptkbp:application |
deep learning
|
| gptkbp:architecture |
massively parallel architecture
|
| gptkbp:collaborates_with |
various universities
AI research labs |
| gptkbp:competes_with |
gptkb:NVIDIA_GPUs
|
| gptkbp:connectivity |
high-speed interconnects
|
| gptkbp:designed_by |
gptkb:Cerebras_Systems
|
| gptkbp:designed_for |
AI model training
AI inference |
| gptkbp:features |
2.6 trillion transistors
|
| gptkbp:form_factor |
single chip
|
| gptkbp:has_impact_on |
gptkb:virtual_reality
gptkb:Telecommunications gptkb:vehicles gptkb:machine_learning gptkb:robotics gptkb:AI_technology image processing computer vision drug discovery financial modeling gaming industry manufacturing processes natural language processing smart cities supply chain optimization data science energy management security systems speech recognition transportation systems climate modeling genomics recommendation systems computational biology social media analysis |
| https://www.w3.org/2000/01/rdf-schema#label |
Wafer Scale Engine 2
|
| gptkbp:manufacturer |
gptkb:TSMC
|
| gptkbp:market |
AI hardware market
|
| gptkbp:notable_feature |
energy efficiency
low latency scalability high throughput extensive memory bandwidth reduced footprint single-chip design |
| gptkbp:performance |
high-performance computing
|
| gptkbp:power_consumption |
high power efficiency
|
| gptkbp:ram |
large on-chip memory
|
| gptkbp:released_in |
gptkb:2021
|
| gptkbp:size |
Wafer-sized
|
| gptkbp:successor |
gptkb:Wafer_Scale_Engine_1
|
| gptkbp:supports |
multiple AI frameworks
|
| gptkbp:target_market |
research institutions
enterprise AI applications |
| gptkbp:technology |
gptkb:AI_technology
|
| gptkbp:uses |
custom silicon
|
| gptkbp:bfsParent |
gptkb:tachymetric_scale
|
| gptkbp:bfsLayer |
5
|