yuria (@hlslyuri) header
yuria (@hlslyuri) profile picture

yuria

San Francisco, CA
Joined November 2020
Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)Photo by yuria (@hlslyuri)
Pinned Tweet
Imagine you find out here that your predicted AWS budget this month is $395k
10
1
157
29,323
i love japanese food
7
58
1,139
yuria retweeted
if ur not a virgin i can change that
61
22
2
495
13,346
i made out with a thai girl drunk and now im getting tweets like this recommended to me
fantasia sexual: estudar em uma escola/universidade só pra mulheres e corromper uma cristã muito devota que chora de culpa religiosa enquanto eu como ela
1
27
663
i want legacy cobol codebases i will buy them!!!!
2
15
467
Never let a boyfriend prevent you from finding your wife
1
9
330
yuria retweeted
ya i’ve been inside of yuria thats how close we are
1
1
8
453
im a tgirl in the wrong body
10
82
1,820
Can't figure out if im lesbian or just a misandrist bisexual
8
1
54
1,690
yuria retweeted
When u break up with someone they start tweeting like sasuke uchiha on here
64
1,287
264
8,692
122,408
i met @wandering_mush but i forgot to bring a second phone so u guys can all just have this photo
3
33
945
yuria retweeted
things to say when ur gf is upset with u 1. i said im sorry like idk what u want me to say
97
2,600
288
30,042
416,208
i miss my ex's mom she was pretty and nice to me
3
54
970
do lesbians make hypergamy worse or better
1
6
210
owning a data company is inherently e-girl coded because you have to like start e-begging for people to want to buy your data and give you money except in real life and like convincingly show your samples are better than the other ones out there except it's data
4
1
1
84
2,244
“The last thing a model sees before being captured in a cuda graph.”
The only thing faster than GB300 Dynamo + vLLM is Vera Rubin Dynamo + vLLM 🏎️💨. The first DeepSeek V4.1 Flash results on @NVIDIA Vera Rubin NVL72 are here, powered by Dynamo + vLLM on @SemiAnalysis_'s AgentX benchmark. Compared with GB300 NVL72 running Dynamo + vLLM, InferenceX's perf ruler shows: • 63.4× total token throughput per chip at 400 tok/s/user • 93.3× at 600 tok/s/user
10
12
3
192
10,465
there's a severe lack of strawberry milk in america
2
7
118
yuria retweeted
anyone else doing the free jensen huang hunger strike? #edtwt
8
7
1
130
2,750
yohji yamamoto bathwater
4
1
27
1,153
this was @0x4132 's
4
264