![RGcloudS's tweet photo. Truth time 👁️
Deep stalking kernel level
Does SMIC really made kirin ?
Does 7nm actually possible with legacy DUV machine ?
Who made Huawei Kirin 9000s ?
Why there's a lot of weird information, like in Image #1 , Kirin 9000s classified as 5nm 😶
Image #2, known information of Hisilicon cpus. kirin 9000 truly listed as 5nm, both 9000 & 9000s recognized as the same chip
Back to Image #1, there's 2 unidentified codes
"72_3330"
"72_3394"
The hexadecimal should be
implementer 0x48 part 0xD02
implementer 0x48 part 0xD42
What does 0xD02 0xD42, etc means?
It's a code set by Arm to name their cores
¹https://t.co/DAK5u05wVB
other cpu parts from different manufacturers also use these kind of code { apple qualcomm broadcom, literally almost everyone }
you can check other names here
[ not updated for 2022+ ]
²https://t.co/e4qpG2OOAQ
So, 0xD42 is Arm A78AE
which is used for automotive & industrial solutions
that makes no sense if kirin use that cpu 💀
Image #3
By the time A78 & A510, huawei already got sanctioned
So they're just guessing next arm core value, A510 was correct prediction
0xD46
But A78 is wrong. arm name it 0xD41
Huawei guessed D42 { A78AE }
How about 0xD02 ?
Well, another suspicious thing
There are 2 cores with same name
As you can check on Link²
{ Number 113 & 114 }
0xD02 is Hisilicon Taishan CPU TSV200 which made almost 3 years ago, first submitted 2 years ago here
https://t.co/k1A5MCLGJN
0xD02 renamed to Linix core 9100 months ago before mate 60 launch
approximately 4 months ago here ↓
https://t.co/KgvyAqBgFa
First of all, you can rename your cpu into anything you want.
But you can't rename 0xD codes, because it's a MIDR_EL1 ID value set by ARM { refer to Link¹ }
MIDR_EL1 register is read-only.
cannot be modified through software means. The value of MIDR_EL1 is stored on the chip eFuse, and eFuse (one-time programmable memory) generally can only refresh data once and cannot be modified twice.
even if the code { 0x } is found to be repeated, the early chips cannot be changed to other values
That's why huawei can only change cpu name but not the value
pretending it was brand new core but not at all.
Apparently huawei did think to name Kirin 9000 to 9100, but the end up with 9100s
Image #4
In Mate60 teardown, 9000s
there's " 2035 "
Means the chip was produced since / in 35th week of 2020
which is 24-30 August 2020
kirin 9000 already in production before arm officially introduced 710 & 510, and huawei guessed it's core value wrong
Summary
Kirin 9000s is actually kirin 9000 made by tsmc 3 years ago
not SMIC
That's why the score of both chips are too similar. obvious
Wait how?
some of you know that chinese has spy everywhere, high patriotism
there's a lot of IP theft news already
{ i don't mean to be racist at all }
Huawei already anticipated this before US issued sanction
So much overstocking
is it really possible to stock that much?
Yes. in 2020, TSMC capacity of combined 6 & 7nm was 140000 wafers per month.
Let's say huawei booked 50%
70000 wafers
1 wafer = around 800 Hisili chips
Tsmc yield is super high that time
Maybe take 85%?
1 wafer = 680 chips
680 x 70000 = 47.600.000 chips per month
In just 3 months, Huawei can overstock 142 million chips of combined Kirin 9000E & 9000
SMIC 7 N+2 yield was too low { <20% }
Mate 60 was too good & smooth
Even samsung at 35% yield on gen 1 & exynos 2200 was so trash
It doesn't matter how talented you are, you can't produce 7nm with legacy 1980 DUV machine, multiple steps of stacking to create compared to EUV
Especially huawei targets 40 million of mate60 sales with such yield?
nonsense
I'm just wondering, why did huawei so pessimistic losing the battle on court and decide to overstock tons same chip? are they really spying ?
well. who knows
Everybody spies anyway, doesn't matter who you were
Just like Ireland decide to keep huawei 5G, because both sides are spying
Oh well, forget politics
Huawei & SMIC successfully pranked the whole world 🫠](https://pbs.twimg.com/media/F8EKAX8asAA_O38.jpg)
![RGcloudS's tweet photo. Truth time 👁️
Deep stalking kernel level
Does SMIC really made kirin ?
Does 7nm actually possible with legacy DUV machine ?
Who made Huawei Kirin 9000s ?
Why there's a lot of weird information, like in Image #1 , Kirin 9000s classified as 5nm 😶
Image #2, known information of Hisilicon cpus. kirin 9000 truly listed as 5nm, both 9000 & 9000s recognized as the same chip
Back to Image #1, there's 2 unidentified codes
"72_3330"
"72_3394"
The hexadecimal should be
implementer 0x48 part 0xD02
implementer 0x48 part 0xD42
What does 0xD02 0xD42, etc means?
It's a code set by Arm to name their cores
¹https://t.co/DAK5u05wVB
other cpu parts from different manufacturers also use these kind of code { apple qualcomm broadcom, literally almost everyone }
you can check other names here
[ not updated for 2022+ ]
²https://t.co/e4qpG2OOAQ
So, 0xD42 is Arm A78AE
which is used for automotive & industrial solutions
that makes no sense if kirin use that cpu 💀
Image #3
By the time A78 & A510, huawei already got sanctioned
So they're just guessing next arm core value, A510 was correct prediction
0xD46
But A78 is wrong. arm name it 0xD41
Huawei guessed D42 { A78AE }
How about 0xD02 ?
Well, another suspicious thing
There are 2 cores with same name
As you can check on Link²
{ Number 113 & 114 }
0xD02 is Hisilicon Taishan CPU TSV200 which made almost 3 years ago, first submitted 2 years ago here
https://t.co/k1A5MCLGJN
0xD02 renamed to Linix core 9100 months ago before mate 60 launch
approximately 4 months ago here ↓
https://t.co/KgvyAqBgFa
First of all, you can rename your cpu into anything you want.
But you can't rename 0xD codes, because it's a MIDR_EL1 ID value set by ARM { refer to Link¹ }
MIDR_EL1 register is read-only.
cannot be modified through software means. The value of MIDR_EL1 is stored on the chip eFuse, and eFuse (one-time programmable memory) generally can only refresh data once and cannot be modified twice.
even if the code { 0x } is found to be repeated, the early chips cannot be changed to other values
That's why huawei can only change cpu name but not the value
pretending it was brand new core but not at all.
Apparently huawei did think to name Kirin 9000 to 9100, but the end up with 9100s
Image #4
In Mate60 teardown, 9000s
there's " 2035 "
Means the chip was produced since / in 35th week of 2020
which is 24-30 August 2020
kirin 9000 already in production before arm officially introduced 710 & 510, and huawei guessed it's core value wrong
Summary
Kirin 9000s is actually kirin 9000 made by tsmc 3 years ago
not SMIC
That's why the score of both chips are too similar. obvious
Wait how?
some of you know that chinese has spy everywhere, high patriotism
there's a lot of IP theft news already
{ i don't mean to be racist at all }
Huawei already anticipated this before US issued sanction
So much overstocking
is it really possible to stock that much?
Yes. in 2020, TSMC capacity of combined 6 & 7nm was 140000 wafers per month.
Let's say huawei booked 50%
70000 wafers
1 wafer = around 800 Hisili chips
Tsmc yield is super high that time
Maybe take 85%?
1 wafer = 680 chips
680 x 70000 = 47.600.000 chips per month
In just 3 months, Huawei can overstock 142 million chips of combined Kirin 9000E & 9000
SMIC 7 N+2 yield was too low { <20% }
Mate 60 was too good & smooth
Even samsung at 35% yield on gen 1 & exynos 2200 was so trash
It doesn't matter how talented you are, you can't produce 7nm with legacy 1980 DUV machine, multiple steps of stacking to create compared to EUV
Especially huawei targets 40 million of mate60 sales with such yield?
nonsense
I'm just wondering, why did huawei so pessimistic losing the battle on court and decide to overstock tons same chip? are they really spying ?
well. who knows
Everybody spies anyway, doesn't matter who you were
Just like Ireland decide to keep huawei 5G, because both sides are spying
Oh well, forget politics
Huawei & SMIC successfully pranked the whole world 🫠](https://pbs.twimg.com/media/F8EJ_4ha0AAba54.jpg)
![RGcloudS's tweet photo. Truth time 👁️
Deep stalking kernel level
Does SMIC really made kirin ?
Does 7nm actually possible with legacy DUV machine ?
Who made Huawei Kirin 9000s ?
Why there's a lot of weird information, like in Image #1 , Kirin 9000s classified as 5nm 😶
Image #2, known information of Hisilicon cpus. kirin 9000 truly listed as 5nm, both 9000 & 9000s recognized as the same chip
Back to Image #1, there's 2 unidentified codes
"72_3330"
"72_3394"
The hexadecimal should be
implementer 0x48 part 0xD02
implementer 0x48 part 0xD42
What does 0xD02 0xD42, etc means?
It's a code set by Arm to name their cores
¹https://t.co/DAK5u05wVB
other cpu parts from different manufacturers also use these kind of code { apple qualcomm broadcom, literally almost everyone }
you can check other names here
[ not updated for 2022+ ]
²https://t.co/e4qpG2OOAQ
So, 0xD42 is Arm A78AE
which is used for automotive & industrial solutions
that makes no sense if kirin use that cpu 💀
Image #3
By the time A78 & A510, huawei already got sanctioned
So they're just guessing next arm core value, A510 was correct prediction
0xD46
But A78 is wrong. arm name it 0xD41
Huawei guessed D42 { A78AE }
How about 0xD02 ?
Well, another suspicious thing
There are 2 cores with same name
As you can check on Link²
{ Number 113 & 114 }
0xD02 is Hisilicon Taishan CPU TSV200 which made almost 3 years ago, first submitted 2 years ago here
https://t.co/k1A5MCLGJN
0xD02 renamed to Linix core 9100 months ago before mate 60 launch
approximately 4 months ago here ↓
https://t.co/KgvyAqBgFa
First of all, you can rename your cpu into anything you want.
But you can't rename 0xD codes, because it's a MIDR_EL1 ID value set by ARM { refer to Link¹ }
MIDR_EL1 register is read-only.
cannot be modified through software means. The value of MIDR_EL1 is stored on the chip eFuse, and eFuse (one-time programmable memory) generally can only refresh data once and cannot be modified twice.
even if the code { 0x } is found to be repeated, the early chips cannot be changed to other values
That's why huawei can only change cpu name but not the value
pretending it was brand new core but not at all.
Apparently huawei did think to name Kirin 9000 to 9100, but the end up with 9100s
Image #4
In Mate60 teardown, 9000s
there's " 2035 "
Means the chip was produced since / in 35th week of 2020
which is 24-30 August 2020
kirin 9000 already in production before arm officially introduced 710 & 510, and huawei guessed it's core value wrong
Summary
Kirin 9000s is actually kirin 9000 made by tsmc 3 years ago
not SMIC
That's why the score of both chips are too similar. obvious
Wait how?
some of you know that chinese has spy everywhere, high patriotism
there's a lot of IP theft news already
{ i don't mean to be racist at all }
Huawei already anticipated this before US issued sanction
So much overstocking
is it really possible to stock that much?
Yes. in 2020, TSMC capacity of combined 6 & 7nm was 140000 wafers per month.
Let's say huawei booked 50%
70000 wafers
1 wafer = around 800 Hisili chips
Tsmc yield is super high that time
Maybe take 85%?
1 wafer = 680 chips
680 x 70000 = 47.600.000 chips per month
In just 3 months, Huawei can overstock 142 million chips of combined Kirin 9000E & 9000
SMIC 7 N+2 yield was too low { <20% }
Mate 60 was too good & smooth
Even samsung at 35% yield on gen 1 & exynos 2200 was so trash
It doesn't matter how talented you are, you can't produce 7nm with legacy 1980 DUV machine, multiple steps of stacking to create compared to EUV
Especially huawei targets 40 million of mate60 sales with such yield?
nonsense
I'm just wondering, why did huawei so pessimistic losing the battle on court and decide to overstock tons same chip? are they really spying ?
well. who knows
Everybody spies anyway, doesn't matter who you were
Just like Ireland decide to keep huawei 5G, because both sides are spying
Oh well, forget politics
Huawei & SMIC successfully pranked the whole world 🫠](https://pbs.twimg.com/media/F8EJ-auacAAiqtJ.jpg)
Olivia
Online
L. Tanawut
@ongloveth
中人之姿，各位抬爱了。
Joined August 2014
55 Following
6 Followers
157 Posts
@Marywu1996 @jayravenexe @fukachi0028 @whiteedog_ สุดท้ายแล้ว การกระทำหรือความคิด ของคุณก็ไม่แตกต่างจากสิ่งที่ผมพูดและอธิบายไปในคอมเมนต์ก่อนหน้านี้เลย ซึ่งผมก็ไม่แปลกใจเลย
@Marywu1996 @jayravenexe @fukachi0028 @whiteedog_ พวกเขามีล่ามในการคุย และ ผู้หญิงได้ถูกเชิญออก แต่เธอเลือกที่จะทำหูทวนลม ไม่ฟัง ก็ไม่แปลกใจที่เจ้าหน้าที่จะกระทำกับเธอแบบนั้น ในประเทศของคุณ เจ้าหน้าที่ก็กระทำแบบนี้ไม่ใช่หรอ ?
@davidmaccartney @visegrad24 So, if the bomb landed at that spot, does that mean the Cambodian people wouldn't blame Thailand?
@LSomala @visegrad24 If the bomb were to drop in that location, would your people blame Thailand or hold your own people responsible?
@cheatakat94 @visegrad24 there's no need to translate anything. Women, children, and the elderly should leave the red zone, not be on the front lines.
The Cambodian Army is using women and children to handle their mortar shells during the current battles with Thai forces
🇹🇭🇰🇭
แผ่นดินไหว ประเทศเมียนมา(21.42°N,96.38°E) ขนาด 7.1 ลึก 10 กม. วันที่ 28 มี.ค. 2568 เวลา 13:32 น. [TMD] https://t.co/QAAH8WRLfl
🚀 April 1st. Open League S1. $115M in $TON for community rewards!
The pilot season was just a teaser – time to double down on fun & rewards with the first full season, starting Apr 1st.
Want to be part of one of the biggest Web3 events ever? This 🧵 is for you 👇
We've found the issue!
Please give us some time to resolve it.
Problem solved.
You may leave this shitty app now.
Enjoy.😁🫶
Priscilla telling me not to post anything here again.
Let's go back to the other side, friends. 😬😬
Truth time 👁️
Deep stalking kernel level
Does SMIC really made kirin ?
Does 7nm actually possible with legacy DUV machine ?
Who made Huawei Kirin 9000s ?
Why there's a lot of weird information, like in Image #1 , Kirin 9000s classified as 5nm 😶
Image #2, known information of Hisilicon cpus. kirin 9000 truly listed as 5nm, both 9000 & 9000s recognized as the same chip
Back to Image #1, there's 2 unidentified codes
"72_3330"
"72_3394"
The hexadecimal should be
implementer 0x48 part 0xD02
implementer 0x48 part 0xD42
What does 0xD02 0xD42, etc means?
It's a code set by Arm to name their cores
¹https://t.co/DAK5u05wVB
other cpu parts from different manufacturers also use these kind of code { apple qualcomm broadcom, literally almost everyone }
you can check other names here
[ not updated for 2022+ ]
²https://t.co/e4qpG2OOAQ
So, 0xD42 is Arm A78AE
which is used for automotive & industrial solutions
that makes no sense if kirin use that cpu 💀
Image #3
By the time A78 & A510, huawei already got sanctioned
So they're just guessing next arm core value, A510 was correct prediction
0xD46
But A78 is wrong. arm name it 0xD41
Huawei guessed D42 { A78AE }
How about 0xD02 ?
Well, another suspicious thing
There are 2 cores with same name
As you can check on Link²
{ Number 113 & 114 }
0xD02 is Hisilicon Taishan CPU TSV200 which made almost 3 years ago, first submitted 2 years ago here
https://t.co/k1A5MCLGJN
0xD02 renamed to Linix core 9100 months ago before mate 60 launch
approximately 4 months ago here ↓
https://t.co/KgvyAqBgFa
First of all, you can rename your cpu into anything you want.
But you can't rename 0xD codes, because it's a MIDR_EL1 ID value set by ARM { refer to Link¹ }
MIDR_EL1 register is read-only.
cannot be modified through software means. The value of MIDR_EL1 is stored on the chip eFuse, and eFuse (one-time programmable memory) generally can only refresh data once and cannot be modified twice.
even if the code { 0x } is found to be repeated, the early chips cannot be changed to other values
That's why huawei can only change cpu name but not the value
pretending it was brand new core but not at all.
Apparently huawei did think to name Kirin 9000 to 9100, but the end up with 9100s
Image #4
In Mate60 teardown, 9000s
there's " 2035 "
Means the chip was produced since / in 35th week of 2020
which is 24-30 August 2020
kirin 9000 already in production before arm officially introduced 710 & 510, and huawei guessed it's core value wrong
Summary
Kirin 9000s is actually kirin 9000 made by tsmc 3 years ago
not SMIC
That's why the score of both chips are too similar. obvious
Wait how?
some of you know that chinese has spy everywhere, high patriotism
there's a lot of IP theft news already
{ i don't mean to be racist at all }
Huawei already anticipated this before US issued sanction
So much overstocking
is it really possible to stock that much?
Yes. in 2020, TSMC capacity of combined 6 & 7nm was 140000 wafers per month.
Let's say huawei booked 50%
70000 wafers
1 wafer = around 800 Hisili chips
Tsmc yield is super high that time
Maybe take 85%?
1 wafer = 680 chips
680 x 70000 = 47.600.000 chips per month
In just 3 months, Huawei can overstock 142 million chips of combined Kirin 9000E & 9000
SMIC 7 N+2 yield was too low { <20% }
Mate 60 was too good & smooth
Even samsung at 35% yield on gen 1 & exynos 2200 was so trash
It doesn't matter how talented you are, you can't produce 7nm with legacy 1980 DUV machine, multiple steps of stacking to create compared to EUV
Especially huawei targets 40 million of mate60 sales with such yield?
nonsense
I'm just wondering, why did huawei so pessimistic losing the battle on court and decide to overstock tons same chip? are they really spying ?
well. who knows
Everybody spies anyway, doesn't matter who you were
Just like Ireland decide to keep huawei 5G, because both sides are spying
Oh well, forget politics
Huawei & SMIC successfully pranked the whole world 🫠
![RGcloudS's tweet photo. Truth time 👁️
Deep stalking kernel level
Does SMIC really made kirin ?
Does 7nm actually possible with legacy DUV machine ?
Who made Huawei Kirin 9000s ?
Why there's a lot of weird information, like in Image #1 , Kirin 9000s classified as 5nm 😶
Image #2, known information of Hisilicon cpus. kirin 9000 truly listed as 5nm, both 9000 & 9000s recognized as the same chip
Back to Image #1, there's 2 unidentified codes
"72_3330"
"72_3394"
The hexadecimal should be
implementer 0x48 part 0xD02
implementer 0x48 part 0xD42
What does 0xD02 0xD42, etc means?
It's a code set by Arm to name their cores
¹https://t.co/DAK5u05wVB
other cpu parts from different manufacturers also use these kind of code { apple qualcomm broadcom, literally almost everyone }
you can check other names here
[ not updated for 2022+ ]
²https://t.co/e4qpG2OOAQ
So, 0xD42 is Arm A78AE
which is used for automotive & industrial solutions
that makes no sense if kirin use that cpu 💀
Image #3
By the time A78 & A510, huawei already got sanctioned
So they're just guessing next arm core value, A510 was correct prediction
0xD46
But A78 is wrong. arm name it 0xD41
Huawei guessed D42 { A78AE }
How about 0xD02 ?
Well, another suspicious thing
There are 2 cores with same name
As you can check on Link²
{ Number 113 & 114 }
0xD02 is Hisilicon Taishan CPU TSV200 which made almost 3 years ago, first submitted 2 years ago here
https://t.co/k1A5MCLGJN
0xD02 renamed to Linix core 9100 months ago before mate 60 launch
approximately 4 months ago here ↓
https://t.co/KgvyAqBgFa
First of all, you can rename your cpu into anything you want.
But you can't rename 0xD codes, because it's a MIDR_EL1 ID value set by ARM { refer to Link¹ }
MIDR_EL1 register is read-only.
cannot be modified through software means. The value of MIDR_EL1 is stored on the chip eFuse, and eFuse (one-time programmable memory) generally can only refresh data once and cannot be modified twice.
even if the code { 0x } is found to be repeated, the early chips cannot be changed to other values
That's why huawei can only change cpu name but not the value
pretending it was brand new core but not at all.
Apparently huawei did think to name Kirin 9000 to 9100, but the end up with 9100s
Image #4
In Mate60 teardown, 9000s
there's " 2035 "
Means the chip was produced since / in 35th week of 2020
which is 24-30 August 2020
kirin 9000 already in production before arm officially introduced 710 & 510, and huawei guessed it's core value wrong
Summary
Kirin 9000s is actually kirin 9000 made by tsmc 3 years ago
not SMIC
That's why the score of both chips are too similar. obvious
Wait how?
some of you know that chinese has spy everywhere, high patriotism
there's a lot of IP theft news already
{ i don't mean to be racist at all }
Huawei already anticipated this before US issued sanction
So much overstocking
is it really possible to stock that much?
Yes. in 2020, TSMC capacity of combined 6 & 7nm was 140000 wafers per month.
Let's say huawei booked 50%
70000 wafers
1 wafer = around 800 Hisili chips
Tsmc yield is super high that time
Maybe take 85%?
1 wafer = 680 chips
680 x 70000 = 47.600.000 chips per month
In just 3 months, Huawei can overstock 142 million chips of combined Kirin 9000E & 9000
SMIC 7 N+2 yield was too low { <20% }
Mate 60 was too good & smooth
Even samsung at 35% yield on gen 1 & exynos 2200 was so trash
It doesn't matter how talented you are, you can't produce 7nm with legacy 1980 DUV machine, multiple steps of stacking to create compared to EUV
Especially huawei targets 40 million of mate60 sales with such yield?
nonsense
I'm just wondering, why did huawei so pessimistic losing the battle on court and decide to overstock tons same chip? are they really spying ?
well. who knows
Everybody spies anyway, doesn't matter who you were
Just like Ireland decide to keep huawei 5G, because both sides are spying
Oh well, forget politics
Huawei & SMIC successfully pranked the whole world 🫠](https://pbs.twimg.com/media/F8EKA_TbUAAvYlX.jpg)
ระแวงฉิXหายเลย !!! สรยุทธ ถึงเอ่ยปากระแวงกันฉิxหาย หลัง #วิโรจน์ เปิดข้อมูลลูกสมุนโทรหานายแต่ไม่รับ สุดท้ายเลขาโทรมาตัดหาง บอกอย่าปากโป้ง เตือนแล้วอย่าทำไม่มีใครมาดูแล แต่ลูกสมุนก็เก็บหลักฐานเอาไว้หมดเหมือนกัน ตอนนี้ระแวงกันหมดแล้ว
#หุ้นitv
#กรรมกรข่าวคุยนอกจอ
#ข่าว3มิติ
ตอนนี้นรกอยู่ในใจ ระแวงไปหมด ใจหนึ่งก็กลัวติดคุก อีกใจหนึ่งก็รักชีวิตกลัวถูกเก็บ
Last Seen Users on Sotwe
เอ้าดอร์เสียว คู่👬
Seen from Thailand
trai đâm tìm chị lớn tuổi
Seen from Vietnam
Iain Kirtley
hitomi tanaka fans
Seen from Vietnam
Tobrut ajah
Seen from Indonesia
evli çift istanbul
Seen from Turkey
hahawewe
Seen from Indonesia
⭐ 𝐆𝐎𝐎𝐍𝐄𝐑 𝐏𝐎𝐑𝐍 𝐒𝐓𝐀𝐑𝐒 ⭐
emrearzu-emrearzu3330
Seen from Turkey
Mostafa Qwer1234
Seen from Egypt
Trends for you
Most Popular Users
Elon Musk 
@elonmusk
241M followers
Barack Obama
@barackobama
119.2M followers
Cristiano Ronaldo
@cristiano
112.3M followers
Donald J. Trump
@realdonaldtrump
111.8M followers
Narendra Modi
@narendramodi
107.1M followers
Rihanna
@rihanna
98.1M followers
NASA
@nasa
92.2M followers
Justin Bieber
@justinbieber
91.3M followers
KATY PERRY
@katyperry
88.6M followers
Taylor Swift
@taylorswift13
82.5M followers
Lady Gaga
@ladygaga
74M followers
Virat Kohli
@imvkohli
71.4M followers
Kim Kardashian
@kimkardashian
70.2M followers
YouTube
@youtube
68.7M followers
Bill Gates
@billgates
64.4M followers
Neymar Jr
@neymarjr
64.2M followers
The Ellen Show
@theellenshow
62.4M followers
CNN
@cnn
61.8M followers
Selena Gomez
@selenagomez
61.7M followers
X
@x
60.8M followers