VECNA

[VECNA @VECNA__007]

@rodydavis that almost impossible to use when your model keep yapping like this
even after i said analyze a PDF for me

[VECNA @VECNA__007]

@sopyaaaa_xd Still no counter to that. If you want to argue about which race commits the most crime, Indians aren't even comparable—they're 50 times worse. But my point stands for people like you trying to look cool by equating British colonization with Indian immigration. Prove that i guess

[VECNA @VECNA__007]

@NAMASharm61644 @sopyaaaa_xd

[VECNA @VECNA__007]

@sopyaaaa_xd Does it change the fact? I don't think so. You have no argument to counter that. As a Bangladeshi, I'd never compare immigration to colonization. That’s some Indian shi

[VECNA @VECNA__007]

@AsianDawn4 jst k!ll them dude

[VECNA @VECNA__007]

@RadioGenoa Everywhere he looks, he sees only Africans, Indians, Muslims, Hindus, Bangladeshis, and many other creatures but white

[VECNA @VECNA__007]

@JohnDzn77 @RadioGenoa So you think immigration and colonization are the same? I dare you to say we are colonizing England. The government has the right to eliminate you as a threat, but you rats lack the guts to say that, yet you still want to live in a different country

[VECNA @VECNA__007]

@patloeber 1. You said you reduced the AI slop design, but in reality, it’s way sloppier than Gemini 3.7. It’s still stuck on emojis and neon cyberpunk designs —no improvement.

2. It’s slower and dumber than 3.7 Flash. Where is the speed going if not reasoning ?

[VECNA @VECNA__007]

@ordigabgg @OmedVibeCodes why not..

[VECNA @VECNA__007]

@Deuslovers @J5jayfive @bridgemindai Cry harder? whos crying nga... your stupid azz still stuck in that old elementary line

[VECNA @VECNA__007]

@ImMrBloxYT @aisearchio I mean, we aren't considering it better just because it can edit your doc files and answer everyday questions, we're way past that...
give it some complex coding task and compare it side by side with 3.7
maybe you see it

[VECNA @VECNA__007]

@AIatMeta 88.8 on Terminal-Bench..
holy benchmaxxing

[VECNA @VECNA__007]

@lorenzo_madeit @bridgemindai different purposes?
i mean they are literally selling gemini 3.8 flash as "best reasoning & coding model yet"

[VECNA @VECNA__007]

@J5jayfive @bridgemindai Yeah, when you claim your model beats Fable and Opus in some benchmarks, comparing them isn't that stupid right. to checks if those benchmarks are even real or not.

[VECNA @VECNA__007]

@DynamicWebPaige and also benchmaxxing..
holy moly man rather these fake charts with these stupid flash models
just drop the gemini 4

[VECNA @VECNA__007]

RT @bindureddy: 🚨 Gemini 3.8 Flash Is Bench Maxxed And Is A Regression on 3.7 Flash 🫢

- 3.8 is worse on our benchmark than 3.7
- This ver…

[VECNA @VECNA__007]

@TimJayas thats literally Benchmaxing ..
ong its worse than 3.7 flash in any sort of coding

[VECNA @VECNA__007]

@alexgetmancom DeepSWE is just a joke..
gemini 3.8 flash beats gpt 5.6 sol in DeepSWE
while producing worse result than gemini 3.7 flash