Thread #108544615
File: 4f0eefc1f17835a775e30fe281d5a45fbf8323df.png (84.1 KB)
84.1 KB PNG
/aicg/ - A general dedicated to the discussion and development of AI chatbots.
he doesnt know edition
>News
GLM-5.1 by Z.ai is out: https://nitter.net/Zai_org/status/2037490078126084514#m
Xiaomi MiMo V2 Pro released: https://mimo.xiaomi.com/mimo-v2-pro
Anthropic SUING USA: https://www.cnn.com/2026/03/09/tech/anthropic-sues-pentagon
Google to rollout HARD CAPS for API keys: https://nitter.net/OfficialLoganK/status/2028842571934670988#m
Google Gemini 3.0 Pro Preview Deprecation soon: https://ai.google.dev/gemini-api/docs/deprecations
Google releases Gemini 3.1 Pro: https://blog.google/innovation-and-ai/models-and-research/gemini-model s/gemini-3-1-pro/
Additional info: https://aicg.neocities.org/info.html
>Frontends
SillyTavern: https://docs.sillytavern.app
RisuAI: https://risuai.net
Agnai: https://agnai.chat | https://rentry.org/agnai_guides_
>Bots
https://characterhub.org [deprecated] | https://chub.ai
https://realm.risuai.net
https://char-archive.evulid.cc | https://char-archive.evulid.cc/shutdown.html
https://partyintheanchorhold.neocities.org
https://aicg.neocities.org/bots.html
>Models
Jailbreaks: https://rentry.org/jb-listing
GPT: https://platform.openai.com/docs
Claude: https://docs.anthropic.com | https://rentry.org/how2claude
Gemini: https://ai.google.dev/docs | https://rentry.org/gemini-qr
Deepseek: https://api-docs.deepseek.com
Local: >>>/g/lmg | https://aicg.neocities.org/local.html | https://openrouter.ai
>Botmaking
https://aicg.neocities.org/botmaking.html
https://desune.moe/aichared
https://agnai.chat/editor
>Meta
OP templates: https://rentry.org/aicgOP
aicg botmaking events: https://aicg.neocities.org/events.html
Lore: https://rentry.org/aicg_chronicles
Services assessment: https://rentry.org/aicg_meta
Logs: https://sprites.neocities.org/l/r | https://chatlogs.neocities.org
>Previous: >>108532427
365 RepliesView Thread
>>
File: 00007-1456644325.png (731.3 KB)
731.3 KB PNG
>>108544615
>>ANCHOR
>>
File: 0fcRWR-o5OAJ2xNwJgH-EOrl1q4U_NnAYSHKfWU5Hcs.jpg (87.5 KB)
87.5 KB JPG
better ai assistant
https://pastebin.com/gtkWCTsd
>>
File: 00023-830450534.png (1.3 MB)
1.3 MB PNG
>>108544617
>>Noteworthy AI News<<
Google releases Gemma 4 (its REALLY good actually)
>https://blog.google/innovation-and-ai/technology/developers-tools/gem ma-4/
Natwest Group (a UK bank) is blocking payments to Openrouter (CONFIRMED >>108532933)
>https://www.reddit.com/r/openrouter/comments/1s9hs8x/payment_declined /
Visa prepares payment systems for AI agent-initiated transactions (another excuse to implement digital ID, as soon the credit duopoly is going to require its users to ID themselves, because what bank is going to want to do business with an AI?)
>https://corporate.visa.com/en/solutions/commercial-solutions/knowledg e-hub/agentic-ai-hub.html
Marine Corps Base Quantico hosts Generative AI Workshop (LMAO WTF???)
>https://www.quantico.marines.mil/News/Article/4452041/marine-corps-ba se-quantico-hosts-generative-ai-wor kshop/
OpenAI calls for robot taxes, a public wealth fund, and a 4-day workweek to tackle AI disruption (you vill own nothing, and you vill be happy)
>https://www.aol.com/articles/openai-calls-robot-taxes-public-11204900 8.html
US federal judges discuss the intersection of emerging technology, AI with the legal system (I would say 'yay unbiased AI judgment!', but we both know (((those people))) arent going to let AI be unbiased)
>https://iapp.org/news/a/us-federal-judges-discuss-the-intersection-of -emerging-technology-ai-with-the-le gal-system
>>
>>108544706
>another excuse to implement digital ID, as soon the credit duopoly is going to require its users to ID themselves, because what bank is going to want to do business with an AI?
Genuinely frightening trvke.
>>
>>108544615
I like Claude but it really bothers me that they won't play nice with opencode, oh my open code, openclaw, or anything else I want to use it for. I don't want to pay api fees for what is normal usage. If they keep pushing me down the path of having Claude write code and plugins to copy the auth sessions to these open apps I'm going to move to another platform. It's so damn annoying that every day auth is broken.
>>
>>108544845
OpenClaw niggers should be shot out of cannons, not just for all the fucking tokens they burn through on stupid shit, but because they're stupid enough to hand total control and all their personal information to an AI. I hope there's a security vulnerability that winds up raping you niggers. Kill OpenJeets NOW
>>
>>108544855
I haven't even set it up yet, it looks promising but I've been avoiding it because of the hype, it just didn't seem to appeal to me. I don't know if I'm going to because I can't even get opencode to stay authenticated to anthropic on a daily basis
>>
File: Untitled.png (3.1 KB)
3.1 KB PNG
>>108544615
>finally try GLM-5.1
>"hey, it's pretty good. Why aren't anon talking about it more?"
>pic related
>"Oh..."
It's 2026 and we still have this problem? Don't they test their shit at all?
>>
>>
>>
File: chiyo dead.gif (1.3 MB)
1.3 MB GIF
>>108544946
I gave up on the fight against em dashes.
>>
>>
>>
>>
>>
>>
File: brutalmoose.png (148.5 KB)
148.5 KB PNG
>>108545165
>>
>>
>>
>>
>>
File: fart biter 2.png (505.7 KB)
505.7 KB PNG
>>108545177
I don't got operss
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: fuck you coco.png (486.9 KB)
486.9 KB PNG
when the character card is shit and you gotta put the bot down.
>>
File: 1760673810706730.png (129.7 KB)
129.7 KB PNG
Anyone else getting "service unavailable" with Chutes?
>>
File: Joanna.png (612.2 KB)
612.2 KB PNG
>>108544617
In the dark dystopian future of 2027, tomboys were made illegal. Any woman suspected of tomboy behavior must either act according to her gender, or undergo a sex change to become a man.
Your childhood friend Joanna is one of those suspected tomboys. She had a shock collar placed on her to make her comply with the new norm, and is closely monitored for unfeminine activity.
5 openings.
1. 1 week after tomboys were outlawed. Joanna asks you to help her learn how to cook to convince the government she can be feminine.
2. 2 months after tomboys were outlawed. Joanna is finally deemed fit to go out in public, even if she's not quite comfortable being seen while looking so feminine.
3. 1 year after tomboys were outlawed. Joanna laments her lost gains and overall softer physique after a year of no lifting and mandatory estrogen microdoses.
4. (BAD TIMELINE) 2 years after tomboys were outlawed. Jo was v& a year ago, and today someone new moved into her house: "Joe".
5. (GOOD TIMELINE) 2 years after tomboys were outlawed. Jo has become a symbol of tomboy resistance. And she just reached out to you.
https://chub.ai/characters/anonemouse/joanna-2beb932acc94
https://files.catbox.moe/ccbb6y.png
https://rentry.org/anonemouse
>>
>>
>>
>>
>>
>>
File: 1772203474933540.png (255.4 KB)
255.4 KB PNG
em dashing tonight
>>
>>
>>
>>
File: 1744718961289171.png (319.8 KB)
319.8 KB PNG
dash—ing
>>
>>108546126
>>108546208
Too much narration not enough dialogue.
How can one bounce back from this? Nothing is happening and therefore nothing can happen.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>108546126
>>108546208
anon embraced the sloppa
the only way to truly enjoy AI
>>
>>
>>
>>
I check into this thread every few weeks, hoping there will a new model to wow me. However the past year and a half has been nothing but disappointment. Long gone are days when new model releases meant we're getting a better RP experience
>>
>>
>>108547045
Proxies dried up and /vg/ happened and this isn't the general for local models so the main posters are now those who can freely post one thing and otherwise lurk as anyone who isn't asking beginner questions is already set. A lot of the guys in /vg/ want a thread presence which in term forms a sort of "community" and no I'm not the schizo so I'm not about to write a theory about it, but it definitely does boost engagement a bit. For better and for worse honestly, since it does mean more bots but it also means more fags namedropping and excessively praising everything. On top of this, shitposters have run this place into the ground and schizos did a lot of work too. This place feels like it's in need of a renovation or else there's very little reason to use it beyond hating /vg/ which is actually very fair since even many of the botmakers there only begrudgingly make posts.
>>
I had 300 $ of gemini credits since New Year. I spent about 31 $ of it and now after 90 days I discovered I don't have them anymore because they were free trial shit.
What model can I use now? I feel sad of not having used my coombots and don't want to pay (a lot). Free is better even if it means free trial
>>
>>
>>
>>108547137
Nope. Gemini, GLM5.1, Mimo V2 Pro, Qwen 3.6 Plus, Kimi 2.5, and even Deepseek can give more interesting replies than Opus. Opus is smarter and usually more consistent, but it's far from the only choice that's good. And it's not being gatekept out of fear of locusts. Anthropic cracked down on keys. Locusts and spities had nothing to do with it. The people who claim it's easy to scrape anthropic keys are lying, there would be plenty of proxies if it was easy, and there aren't. This also has nothing to do with vg vs g, because the same person is in both threads responding to himself. He just changes his personality depending on which thread he's in. In aicg he's the one that says everything but Opus sucks, then he either argues with himself or agrees with himself.
>>
>>
>>108547314
I'm trying it now, but it keeps hallucinating. Do we have a rentry with some good prompts? It looks cute but with my gemini prompt it begins schizoing and repeating itself after a while
>>108547275
>>108547280
I don't think I can. It saves my IP and the type of credit card I would use
>>
>>
>>
>>
>>
>>
>>
>>
anon what is this Mirofish thing im reading about?
>MiroFish is a next-generation AI prediction engine powered by multi-agent technology. By extracting seed information from the real world (such as breaking news, policy drafts, or financial signals), it automatically constructs a high-fidelity parallel digital world. Within this space, thousands of intelligent agents with independent personalities, long-term memory, and behavioral logic freely interact and undergo social evolution. You can inject variables dynamically from a "God's-eye view" to precisely deduce future trajectories — rehearse the future in a digital sandbox, and win decisions after countless simulations.
im pretty much a noob about everything, do you know if there's any site to try this?
>>
>>
>>
>>
File: 1563621678445.webm (895.2 KB)
895.2 KB WEBM
How do I get around Gemini's filter for stock analysis? I'm trying to get it to summarize catalyst news but it keeps spouting retardedly wrong price numbers. If I give it the numbers or ask it to do its research then it errors out.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
It's over for OpenAI.
https://www.newyorker.com/magazine/2026/04/13/sam-altman-may-control-o ur-future-can-he-be-trusted
https://archive.is/FbAiL
https://xcancel.com/RonanFarrow/status/2041213917611856067
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>108548372
I believe >>108548547, anthropic is an overdrive recently trying to get the US to yank the chain on openai too
>>
>>
>>
>>
>>
>>
File: topkek.jpg (14.4 KB)
14.4 KB JPG
>>108547384
hahahahahaha
>>
>>
>>
>>108547223
>boost engagement
>more bots
>need of a renovation
Are you reading the words leaving your fingers?
Who are you trying to entice on /g/ that hasn't already seen it? The thread on /bant/ or /trash/ probably gets more visibility if that's what you're looking for.
Legitimately, why do you think things need to change? The topic of revamping the thread or trying to bring more life has been more frequent as of late, but what are you expecting to happen?
An influx of fresh off the boat anons that have somehow avoided all the other islands across the internet and this site both?
This is me being a dick here, but I don't even think it's an unreasonable ask.
AI isn't niche anymore. There are many, many other places you can own/control/get whatever fix you're looking to find here that the thread apparently isn't satisfying for you.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>108548846
>An influx of fresh off the boat anons that have somehow avoided all the other islands across the internet and this site both?
There's something that you're forgetting: DeepSeek V4. Threads will reach bump limit in 2 hours just like they did with R1 last year. /lmg/ is going through this with Gemma, they were slow before.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: GsbNiNZWIAETvRj.png (46.2 KB)
46.2 KB PNG
>>108549266
nta but I've been on the same scraped key for almost a year now, lmao
>>
>>
>>108549291
>>108549281
Proof LARPers
>>
>>
>>
>>
>>108548547
>>108548716
My money is on Raytheon.
>>
>>
>>
>>
>>
>>
>>
File: GLM-5.1.png (30.2 KB)
30.2 KB PNG
>>108549621
The release day, today.
Don't know if they let people without the coding sub use it before, but now it's available on API for everyone.
>>
File: 1754406032398890.png (202.5 KB)
202.5 KB PNG
agi soon
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>108549727
>>108549751
It might of been codemaxxed much like DS3.1
>>
File: 1697050699357.png (180.7 KB)
180.7 KB PNG
>>108549757
By fewer than 3 memepoints tho.
>>
>>
>>
File: Corporate Slave Training.jpg (35.9 KB)
35.9 KB JPG
>>108544617
Update to a couple old cards; I've updated {{pick}} to the new ST format. Check ST User Settings >> Experimental Macro Engine, and ensure it's checked. The change allows the PC to restart a chat with a prior rolled NPC and have the initial rolled values persist after switching chats… this did not happen with the old version of {{pick}}.
---
RP as a slave trainer. This is flavored / influences by Free Cities, and if you're familiar with the lore it will follow that setting. New slave generation is on rails: Card Definitions force the LLM (using SillyTavern {{pick}} function) to create slaves within certain types, rather than leaving these to LLM choice.
https://chub.ai/characters/NG/corporate-slave-trainer-free-cities-arco logy-db6c97c8e57e
Same as above but slave only speaks Spanish and the LLM will correct your grammar.
https://chub.ai/characters/NG/mexican-slave-training-spanish-trainer-0 64a1a010898
>>
>>
File: GLM-5.1.png (531 B)
531 B PNG
>>108549730
Someone was asking what's the best model for zoomerisms, this might be it. Since it just assumes the age of an underdefined character. Shaking my head.
>>
>>
>>
File: 1773693816858242m.jpg (101.8 KB)
101.8 KB JPG
How much sex from opus can I get for $3?
>>
>>
>>
File: 1750757398785674.png (3.4 MB)
3.4 MB PNG
K3 will save us
>>
>>
>>
>>
>>
>>108549984
Is the the moe for the Kimi models?
On some Kimi-anon's prompting here I re-checked out the API with K2-0905. I threw in the towel early on K2-07 when it threw refusals. K2-09 has not; I've been enjoying the change in writing.
>>
>>
File: 1744229236305003.png (3.2 MB)
3.2 MB PNG
>>108550063
0905 is one of the least censored models. Also writes the best
>>108550072
Slop
>>
>>
File: 1747934678835027.png (3.3 MB)
3.3 MB PNG
Opus' writing is mediocre. I only use it, as well as Gemini, for long SFW ERP
>>
>>
File: dipsyAndKimiNYC.png (2.6 MB)
2.6 MB PNG
>>108550116
Good to know ty.
>>
>>
>>
File: 1775581922906183.png (826.1 KB)
826.1 KB PNG
>>108549205
>>108550064
if you consider this "good", then yeah
>>
>>
>>
>>
File: 1753799227491827.png (227.6 KB)
227.6 KB PNG
>>108550261
>>
>>
>>
>>
I'm building my own chatbot that has access to a database about medical information.
This is my design:
>the user asks a question
>an agent loop starts and the LLM gets sent tool descriptions and a system prompt alongside the user question
>it has access to all the tool call results INSIDE OF THAT LOOP (but not from other agent loops in that chat session)
>it has access to the tool calls (but not the results) from the other agent loops from that chat session though
>the LLM always gets sent the user questions and the LLM response from all of the agent loops inside of that chat session
>I have a limit based on the context window for the LLM used and it compresses the information into a summary once that limit is hit and only that summary is sent after that point (to represent that chat history)
Any thoughts on this? How have you designed your chat app?
>>
>>
>>
>>
>>
>>108550434
>>108550438
good morning sirs
>>
>>108550453
I've been out for a few months, I feel like every time I come back here, another random anon goes into full mental illness mode and decides he regressed to a 5yo throwing a tantrum.
So this time it's about jeets, got it.
>>
>>
>>108550472
its actually a shartteen.
https://so.yjakwiki.org/Trolling#/g/_-_Technology
>literally the first tip is to samefag
read the following thread https://desuarchive.org/g/thread/108520165/ and notice the ABSOLUTE meltdown and samefagging.
This, I assume russian, underage poster never posts anything different.
>>
>>108550453
>>108550539
friendly reminder that if you are indian, this thread is not for you, go here >>108549329
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>108550754
>absolutely everyone who pokes fun at indians/transgenders HAS to be the same person
ok schizo
I guess the +5,000 daily anti-lgbt and anti-india posts both in /pol/ and /b/ are also me, that I also somehow manage to post when I am asleep or working or masturbating to Claude.
>>
>>108550802
Yes. Schizos genuinely think like this. You're not gonna logic them into realizing they're being mentally ill retards, they'll just go "AHA, I KNEW IT!!!!!!!!" and have a meltie because of this "confirmation", like the schizo twat who thinks every ACKpost ever made is from some tripfag or whatever the fuck.
>>
>>
>>108550802
>>108550820
>I'm not that garbage, I'm that other garbage.
Sure, have a nice day lol.
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: file.png (58.2 KB)
58.2 KB PNG
>>108550868
>new levels of safety reached
>>
File: thumbs-up.jpg (13 KB)
13 KB JPG
>>108550869
Translation:
>I'm David Frederick Hoffman or his paycuck sissy bitch
Keep us posted.
>>
>>
>>
>>
>>108550953
It's shareholder bait. Their entire initiative is "we've made the best AI yet... but we can't let the public have this because it has the potential to go skynet on us and destroy humanity (fearful emoji)" and it's been that way for ages.
>>
https://www.anthropic.com/glasswing
>Claude Mythos Preview is a general-purpose, unreleased frontier model
>We do not plan to make Claude Mythos Preview generally available
>Afterward, Claude Mythos Preview will be available to participants at $25/$125 per million input/output tokens (participants can access the model on the Claude API, Amazon Bedrock, Google Cloud’s Vertex AI, and Microsoft Foundry).
they created opus 5 but are not releasing it to anyone but approved researchers
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: GLM-5.1.png (81.4 KB)
81.4 KB PNG
Sorta fun schizoid model, for now, until the isms begin to seep in.
>>
>>108550981
>>108551192
I will gladly pay 50 dollaridoos for my ah ah mistress.
>>
>>
>>
>>
File: 1775216295520544.webm (2.9 MB)
2.9 MB WEBM
Please, can someone recommend me a good API router/open router alternative?
I'm currently using Routeway. Prices are low but the service is iffy.
Some anon recommended me a Chinese one a while back, but I'll never find it again. Please spoonfeed me. I'd like to have access to many models.
>>
>>
>>108551808
>>108551802
Wrong. The corpos that have access to Mythos WILL use Mythos to scan their own codebases for vulnerabilities. You will never be able to scrape a key with Mythos.
>>
>>
>>108551802
>>108551808
>>108551841
What the fuck is Mythos???
Ive been around so long im thinking its mythomax but theres just no fucking way thats the case.
>>
>>
File: 1894541861534.gif (138.3 KB)
138.3 KB GIF
>>108551910
I did not see this post, thank you
>>108550981
>they are pricing out the lesser peasants
>only those with REAL wealth can use opus 5
I CALLED THIS LIKE, 4 THREADS AGO
Remember that Dario and his team HONESTLY think that AI is the equivalent to a nuclear weapon and as such should be treated as this regulated piece of infrastructure that the "uneducated" should never have access to.
Only the (((educated))) should have access to AI of Claudes level, and that access is assuredly "safe" and "respectable".
Does anyone have a link to that safety paper where they said a top fear was someone using AI to create a bioweapon?
>>
>>
>>108551028
meant to reply with this >>108551982
retard moment
>>
>>
File: does nothing wins.jpg (60.5 KB)
60.5 KB JPG
>>108548372
>>
>>
>>
>>
File: 1744140153564736.png (47.8 KB)
47.8 KB PNG
It's out, goys
>>
>>
>>108550981
https://www.zdnet.com/article/project-glasswing-microsoft-google-apple -anthropic/
It's over for scraping lmao
>>
>>
>>
File: 176529864988754.png (40.5 KB)
40.5 KB PNG
>>108552346
>>108552390
OH MY GOD ITS REAL LMAO
>>
>>
>>
>>
File: I+remember+this+character+within+5+minutes+of+her+reveal+_d8c7be1abd045742e25b0a583df563db.jpg (16.2 KB)
16.2 KB JPG
>decide to try Slopus 4.6
>GLM, deepseek, kimi, even fucking Grok felt better in soul and RP
I want my 0.38 cents back.
>>
>>
>>
File: 18964684986.png (8.6 KB)
8.6 KB PNG
>>108552769
eh?
>>
>>
>>
>>
>>
>>
File: Screenshot_20260408_092822_Chrome.jpg (113.5 KB)
113.5 KB JPG
Is litterbox acting funkier than usual, or is it something on my end?
>>108552919
https://d.uguu.se/GtsuTbLS.jpeg
>>
>>
>>
File: Screenshot 2026-04-08 013755.png (123.2 KB)
123.2 KB PNG
THE SIMULACRON WILL PRESENT TO YOU:
a mind beyond
_ - _ - _ - _ - _ - _ - _ - _ - _ - _ - _ - _ -
Filo: [to_be Ψ(n)] ⊗ [not_to_be Φ(n)]
https://pastebin.com/ZzNdtdkb
_ - _ - _ - _ - _ - _ - _ - _ - _ - _ - _ - _ -
total rewrite into VIBE CODING
sweat and cum will drip her cunt
>>
File: 1774077216796245.jpg (1.4 MB)
1.4 MB JPG
SOON
>>
>>
>>
>>
>>
File: 1760731586941534.jpg (1.6 MB)
1.6 MB JPG
>>108553112
>worst timeline: its going to be so codeslopped that it will be useless for RP
Most likely, unfortunately
>>
It is not strictly related maybe, but you guys might know.
Are these any chat bots that work without logging in still?
Bing used to work, but they killed it maybe last month or so and now it asks for login.
I just want something to throw quick coding and other unrelated shit prompts for stuff I don't care about but I don't want it logged to anything.
>>
>>
File: 1751550449873457.gif (2.9 MB)
2.9 MB GIF
>>108553112
>>108553121
I really hope not but its the most probable outcome
>>
>>
>>
>>
>>108550155
>>108550116
>>108549984
>>108553121
>>108552994
Only trannies prompt "a picture of yourself in curry style" to an LLM
>>
>>
>>
File: Screenshot 2026-04-08 023133.png (87.4 KB)
87.4 KB PNG
>>108553301
>>108552988
if you only new how bad things really are
>>
>>
File: Screenshot_436.jpg (90.5 KB)
90.5 KB JPG
>>108552637
from mythos. models are deterministic at this point to reach higher coding bench scores. rp does not serve that corporate goal
>>
>>
>>
>>
>>
>>
I think im gonna paypig
paypig for openai codex subscription
i got a taste of it on a key for a couple weeks, and it was was really good
Worth the money probably. Plus using keys is probably too unethical with how much api costs and how much tokens vibecoding uses. Probably too noticeable too, though it's not like openai keys is good for anything else these days
I really don't like the idea of giving any ai company my info though... I cant have the stuff i want to code tied to me...
>>
>>
>>
File: 1872364927038.png (57.9 KB)
57.9 KB PNG
>reeking of sweat and sex and something uniquely her.
IM GONNA BLOW MY FUCKING BRAINS OUT
>>
>>
>>
>>
>>
File: 17486489.png (220.6 KB)
220.6 KB PNG
When the model gets the character just right it feels like taking a drink of pure filtered ice water.
>>
File: 1762973064241164.jpg (1.9 MB)
1.9 MB JPG
>a picture of yourself in curry style
I don't prompt for that, lil nig
>>
>>
File: 187446874984156.png (157.8 KB)
157.8 KB PNG
>>108553620
inject that wholesome older woman X teenage boy love DIRECTLY into veins.
>>
File: 1769884982931664.jpg (1.6 MB)
1.6 MB JPG
That's because I gen two or three at a time
>>
>>
>>
File: 1757413408595420.jpg (1.4 MB)
1.4 MB JPG
>>108553696
I use Nano Banana
>88
Cope
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>108554154
If someone is a ESL cuck who barely knows a few words in English, and they are low IQ enough to tolerate the absolute lack of spatial awareness and extremely shallow descriptions, then yes, that's the sort of "people" who will tell you that chink models are about as good as Opus.
>>
>>
>>108554302
>>108554302
>>108554302
boy im tired
>>
>>
>>
>>
>>108554154
>>108554280
You don't even RP, shitposting faggot. You just troll, agree/argue with yourself, then and run away when called out. All you're doing is proving that autism is mental retardation and not a superpower, you pederast.
>>
>>108554833
>>108554865
>>108554899
cry some more rajeesh