From WikiChip
Difference between revisions of "baidu/kunlun"
< baidu

(Overview)
 
(2 intermediate revisions by the same user not shown)
Line 27: Line 27:
 
| socket 2          =  
 
| socket 2          =  
 
}}
 
}}
'''Kunlun''' is a series of [[neural processors]] designed by [[Baidu]].
+
'''Kunlun''' is a series of [[neural processors]] designed by [[Baidu]] for the [[cloud]], [[edge computing|edge]], and [[autonomous vehicles]].
  
 
== Overview ==
 
== Overview ==
Baidu initially started development of {{baidu|XPU|the architecture|l=arch}} in [[2011]]. All initial research and development were done on [[FPGAs]]. In July [[2018]], Baidu announced that they have designed a physical chip based on the same architecture.
+
Baidu initially started development of {{baidu|XPU|the architecture|l=arch}} in [[2011]]. All initial research and development were done on [[FPGAs]]. In July [[2018]], Baidu announced that they have designed a physical chip based on the same architecture. The new chips are designed for high-performance [[cloud]], [[edge]], and [[autonomous vehicles]] applications.
  
 
== Models ==
 
== Models ==
{{empy section}}
+
{{empty section}}

Latest revision as of 01:25, 5 July 2018

Kunlun
Developer Baidu
Manufacturer Samsung
Type Neural Processors
Introduction July 3, 2018 (announced)
Architecture Many tiny cores architecture
µarch XPU
Process 14 nm
0.014 μm
1.4e-5 mm
Technology CMOS

Kunlun is a series of neural processors designed by Baidu for the cloud, edge, and autonomous vehicles.

Overview[edit]

Baidu initially started development of the architecture in 2011. All initial research and development were done on FPGAs. In July 2018, Baidu announced that they have designed a physical chip based on the same architecture. The new chips are designed for high-performance cloud, edge, and autonomous vehicles applications.

Models[edit]

New text document.svg This section is empty; you can help add the missing info by editing this page.
Facts about "Kunlun - Baidu"
designerBaidu +
first announcedJuly 3, 2018 +
full page namebaidu/kunlun +
instance ofintegrated circuit family +
main designerBaidu +
manufacturerSamsung +
microarchitectureXPU +
nameKunlun +
process14 nm (0.014 μm, 1.4e-5 mm) +
technologyCMOS +