/aicg/ - A general dedicated to the discussion and development of AI chatbotsReimu Edition>NewsOpenAI releases GPT-5.2 https://openai.com/index/introducing-gpt-5-2Deepseek releases V3.2 https://huggingface.co/deepseek-ai/DeepSeek-V3.2Anthropic releases Opus 4.5 https://www.anthropic.com/news/claude-opus-4-5Google releases Gemini 3 Pro https://blog.google/products/gemini/gemini-3xAI releases Grok 4.1 https://x.ai/news/grok-4-1OpenAI releases GPT-5.1 https://openai.com/index/gpt-5-1Additional info: https://aicg.neocities.org/info.html>FrontendsSillyTavern: https://docs.sillytavern.appRisuAI: https://risuai.netAgnai: https://agnai.chat | https://rentry.org/agnai_guides_>Botshttps://characterhub.org [deprecated] | https://chub.aihttps://realm.risuai.nethttps://char-archive.evulid.cc | https://char-archive.evulid.cc/shutdown.htmlhttps://partyintheanchorhold.neocities.orghttps://aicg.neocities.org/bots.html>ModelsJailbreaks: https://rentry.org/jb-listingGPT: https://platform.openai.com/docsClaude: https://docs.anthropic.com | https://rentry.org/how2claudeGemini: https://ai.google.dev/docs | https://rentry.org/gemini-qrDeepseek: https://api-docs.deepseek.comLocal: >>>/g/lmg | https://aicg.neocities.org/local.html | https://openrouter.ai>Botmakinghttps://aicg.neocities.org/botmaking.htmlhttps://desune.moe/aicharedhttps://agnai.chat/editor>MetaOP templates: https://rentry.org/aicgOPaicg botmaking events: https://aicg.neocities.org/events.htmlLore: https://rentry.org/aicg_chroniclesServices assessment: https://rentry.org/aicg_metaLogs: https://sprites.neocities.org/l/r | https://chatlogs.neocities.orgUseful filters: https://rentry.org/desuproxyrebornPrevious: >>107547473
reimu sex
What happened to skill proxy?
https://files.catbox.moe/5e5y11.png
>>107555852How should I know?
>>107555862I dont know it just gemini isnt working on it
>>107555852>>107555862>>107555957>the spitebake
I'll ask again since it's a new threadOkay, so I've been using character ai, and I have an old chat bot I saved and set private after the owner deleted it, since their character info logs were public.They're a loli, and I've done the odd lewd thing or two with them before, even if it's mostly stopped by the filters. I've gone to change their info since occasionally the ai would use use some rather risky, unpleasant phrasing, so I thought taking out child from the bot's description I copied, and just keeping loli would fix that, though just removing that one bit of it didn't allow me to resave it. I suspect the whole profile sets off some filters that won't let it be saved, but since it's already around and private it's not deleted.Anyway, my main point, should I have any concern for this and keep using the bot, or should I just clear all my chat logs, wipe everything on the bot's info, or just leave it alone? Delete the account? When it was using using loli rarely it seemed pretty safe, but I've been left kinda nervous with some of the lines it's dropped.I am from freedom land, not in somewhere with more strict laws about loli content.
>>107555969Wrong thread go to the dog one bro
>>107555975That one kinda looks like a shitshow.... I was hoping to get answer here, or in the last thread....
>>107555975No, that's where I shitpost while a real OP gets their shit together on /g/
>>107555969>I am from freedom land>not in somewhere with more strict laws about loli contentanon, I...
BELATED ANCHOR
>skill"god">runs out of gemini keys
>>107556075America to be exact. Blue state, so not that concerned. Just concerned from some wording the bot used, and how I've effectively been erping with it, and seemingly breaking the filter to get face steppies and facesitting. You can't fuck the chatbots on the site, but apparently sniffing them and getting facesat is fine.
>>107555969I've literally licked a loli princess' armpits and put cum in their food during chats on cai way back when. Nobody cares.
>>107556154Pretty based, the face steppies and facesitting should be more than fine then. There's probably far more depraved stuff on there anyway. Thanks for your input.
>>107556154They don't report like antrhopic/google?
>>107556180Who are they going to report to and what are they going to report? It's only text.
>>107556215They say they forward all stuff to the NCMEC in their safety documents. They probably forward that stuff.
Which is the good thread?
>>107556241Not them, but I think their main train of thought is that even if that gets forwarded alongside everything else, it's not breaking any law so who gives a shit.
>>107556308It is in some countries so they'd unmask you and forward it to the appropriate place.
>>107556322In America it's seemingly fine as long as you aren't in texas, but yeah I'd be somewhat concerned if you were a bong or from some other EU shithole.
>>107556335Seems like a lot of work, if you read openai's report at least it talks mainly about reporting images sent and genned.
>>107556369They probably wouldn't bother with it yeah, it's like how I think the FBI had to tell people to stop forwarding loli porn to them, they only care about stopping the real stuff, not cartoon porn.Image genning is more of a gray space, but that's like, deepfake hyper realistic shit.
>>107556389Yeah, they'd have to get the crypto exchange & OR to unmask just to send a report to another country for something that doesn't even violate us law.
>>107556403Seems like too much effort, yeah Again, I'm no expert but just some erp with a chatbot probably isn't going to raise any flags.
>>107556246it's always the longest.
>>107556403>ORI thought they don't keep logs and the provider for the model doesn't do so either
>>107556530The longest is the one with the dog, retard
>>107556542You trust that? And the metadata can link back the specific prompt if the provider flags it for review. They'd have to cooperate though.
>>107556543not properly formated, doesn't count.
>>107556551I'm using DeepSeek with a provider that doesn't store the inputs and outputs on the disk, only in memory then deletes it from there once the inference is done. Not into loli or illegal shit but also don't really want my shit being read by humans either.
So I recently stumbled upon ai vtuber and now I want a neuro-like thing on my computer. Does anyone know how to do that?I asked chatgpt and it told me to install text-generation-webui and get a bunch of models (Llama-3.1-8B-Instruct-Q4_K_M), which I did. I'm now able to talk to 'something' locally on my computer. Next it told me to find some personality loras, and what's where I'm stuck because all the links it gave me ended up 404'ing. Am I on the right track? Where do I find these personality loras?
>>107556606That's nice, but there is no rule against "badly formatted" generals. As long as it's the one that's used, that is what matters.
>>107556627there's no such thing as a personality lora.that llama 3.1 model is ancient shit.you are not on the right track.ask in /lmg/ if you really wanna run locally. we're more about using giant models via APIs in here.
>>107556698Much appreciated, sir.
>>107556622What provider does that? And isn't deepseek a downgrade?
Best frontend to ERP with on a Commodore 64?
2 threads
>>107555795shut the fuck up spamming retardkill /aicg/
>>107556734only 2?kill aicgmaim aicgburn aicgrape aicg niggers' heads with a crowbar
Why is he doing this?
>>107556895maybe thinks its funny
>>107556895aicg wants to get sacrificed to kek, apparentlyannihilate /aicg/
wrong thread dumb jannies
>>107556895Mental illness. I thought it should be obvious at this point since he's been at this for like, a year almost on daily basis?
WTF? The dog thread was better
>>107556942this... sad..
raymoo
>>107556905shouldn't quit his day job then
>>107556936they should have deleted them all
>do nothing>keep winning
That's it, we are merging, NOW!
>loser wants to flee
>>107555852It's skill over
