[β]
β
bitchan
[?] help
π
host1.bitchan.one
π
@+U=GT6a7
β±
15h seen
β
0
π
0
β§
1
π hide
send βΏitcoin
[!] follow
[@] review
posts 1
/ 1
[β₯] sort newest
β±
filter [-]
languages:
on
other
/technology
Is Gemini just especially cucked like GPT? I got Grok and Claude Sonnet to write for me.
β±
8h
β©
1
Argon 4 mainly. People get wrong and spam "Skill issue" but Argon itself as a base model is pretty filthy by itself and uncensored. The issue isn't the model but all the classifiers and safety guardrails up in the API. Since google just applied the exact same filters that Google Ai studio had for Vertex at least on Openrouter it makes any sort of roleplay or non-jewish cucked chats nearly-impossible. If you see anyone claiming their replies go through it's likely an proxy that doesn't have these filters or their lying. Generally I just realized no matter how many prompts or jailbreaking advanced or not, the whole thing would be reduced to playing filter whack-a-mole to of avoiding trigger words and context-trigger patterns if your message through. It aren't even worth the effect to play permeant whack-a-mole. Just go back to an older model or just switch. Same reason Opus 4.6 worked because Vertex didn't have the API filters.
β±
8h
β©
1