<s>
A	O
Mobile	B-Architecture
PCI	I-Architecture
Express	I-Architecture
Module	I-Architecture
(	O
MXM	O
)	O
is	O
an	O
interconnect	O
standard	O
for	O
GPUs	B-Architecture
(	O
MXM	B-Architecture
Graphics	I-Architecture
Modules	I-Architecture
)	O
in	O
laptops	B-Device
using	O
PCI	O
Express	O
created	O
by	O
MXM-SIG	O
.	O
</s>
<s>
The	O
goal	O
was	O
to	O
create	O
a	O
non-proprietary	O
,	O
industry	O
standard	O
socket	O
,	O
so	O
one	O
could	O
easily	O
upgrade	O
the	O
graphics	B-Architecture
processor	I-Architecture
in	O
a	O
laptop	B-Device
,	O
without	O
having	O
to	O
buy	O
a	O
whole	O
new	O
system	O
or	O
relying	O
on	O
proprietary	O
vendor	O
upgrades	O
.	O
</s>
<s>
Nvidia	O
GTX	O
980M	O
)	O
"	O
is	O
MXM	O
2.1	O
"	O
,	O
and	O
therefore	O
any	O
notebook	B-Device
with	O
a	O
GTX	O
980M	O
fully	O
implements	O
MXM	O
2.1	O
.	O
</s>
<s>
While	O
Nvidia	O
defines	O
a	O
lot	O
of	O
MXM	O
specifications	O
,	O
they	O
do	O
not	O
manufacture	O
or	O
design	O
MXM	O
cards	O
themselves	O
,	O
which	O
mostly	O
consist	O
of	O
a	O
PCB	O
with	O
vRAM	O
and	O
an	O
Nvidia	O
or	O
AMD	O
GPU	B-Architecture
core	O
.	O
</s>
<s>
Therefore	O
,	O
any	O
model	O
of	O
GPU	B-Architecture
can	O
be	O
manufactured	O
in	O
MXM	O
,	O
but	O
a	O
laptop	B-Device
released	O
with	O
any	O
particular	O
graphics	O
card	O
model	O
may	O
or	O
may	O
not	O
implement	O
MXM	O
regardless	O
.	O
</s>
<s>
The	O
Qseven	B-Device
computer-on-module	B-Application
form	O
factor	O
uses	O
a	O
MXM-II	O
connector	O
,	O
while	O
the	O
SMARC	O
computer-on-module	B-Application
form	O
factor	O
uses	O
a	O
MXM	O
3	O
connector	O
.	O
</s>
