How do I get the model name of my processor?

Question

I am making a conky config and I want to display the CPU model as one of the stats. For this, I need a command that outputs just the CPU name as a string, for example, "Intel Core i7 7700K". I think both /proc/cpuinfo and the output from lscpu have what I need (I would prefer lscpu because it would be slightly more efficient), but the model name is 4 spaces too far away from the manufacturer name. So I would need to cut off the model name: string and get rid of the extra spaces. I would also like to know how to get rid of the @ (clockspeed).

Solution 1

$ lscpu | sed -nr '/Model name/ s/.*:\s*(.*) @ .*/\1/p'
Intel(R) Atom(TM) CPU  Z3735F

Notes

• -n don't print anything until we ask for it
• -r use ERE
• /Model name/ find the line with Model name
• s/old/new/ replace old with new
• .*:\s* anything before a colon, a colon, and any amount of horizontal space
• (.*) @ save any number of characters before ' @'
• \1 reference to saved pattern
• p print only the line we edited

To get rid of double spaces, you can do an extra s command:

$ lscpu | sed -nr '/Model name/ s/  / /g; s/.*:\s*(.*) @ .*/\1/p'
Intel(R) Atom(TM) CPU Z3735F

The doesn't work if there are more than two spaces anywhere, so to delete multiple spaces relentlessly, use a sed loop:

$ lscpu | sed -nr ':a;s/  / /;ta; /Model name/ s/.*: (.*) @ .*/\1/p'
Intel(R) Atom(TM) CPU Z3735F 

This :a;s/ / /;ta keeps on processing the stream until there are no double spaces anywhere. Notice that it won't take an address, so we had to chew through the whole stream before selecting the line. This won't take a noticeable amount of time when we're only parsing the output of lscpu, but on a huge file or on multiple files, it could get pretty slow. We could pipe the output from the original command to another sed with the loop instead to avoid working on the whole stream:

$ lscpu | sed -nr '/Model name/ s/.*:\s*(.*) @ .*/\1/p' | sed ':a;s/  / /;ta'
Intel(R) Atom(TM) CPU Z3735F

Solution 2

lscpu | grep 'Model name' | cut -f 2 -d ":" | awk '{$1=$1}1'

• lscpu To get the CPU information
• grep 'Model name' To extract the line containing CPU name.
• cut -f 2 -d ":" To remove part of the line before :. So it will remove the part "Model name:" from the output.
• awk '{$1=$1}1' To remove the space from the beginning of a line.

Sample Output

$ lscpu | grep 'Model name' | cut -f 2 -d ":" | awk '{$1=$1}1'
Intel(R) Pentium(R) CPU N3520 @ 2.16GHz

Solution 3

Use this to get what you want:

lscpu | grep "Model name:" | sed -r 's/Model name:\s{1,}//g'

Result:

Intel(R) Core(TM) i5-4210U CPU @ 1.70GHz

Or:

lscpu | grep "Model name" | sed -r 's/Model name:\s{1,}(.*) @ .*z\s*/\1/g'

Result:

Intel(R) Core(TM) i5-4210U CPU

Comments

[Haxalicious]

But that doesn't get rid of the "Model name:" at the front.

[George Udosen]

Please post an example of what you want, sorry for the mix up!

[George Udosen]

Doing from mobile so test and let me know

[Haxalicious]

This works, but there is still extra spaces.

[Zanna]

@Haxalicious sorry I was distracted by something. I edited my original version to trim the trailing space, and added a version that removes all double spaces