how to download and use the router to run it on your own computer and let the potato rest as you called it. Thanks for help. 

(overall, I didn't understand much of what you wrote in this manual)

Still trying to get it work somehow, but futile. I guess there is nothing to do about it.

I am looking into running the AI end locally to spare you some resources. Is the API key mandatory within the settings? If I run it only locally with no way for it to leave my device, it shouldn't matter too much who can use it to be honest.

What if, when selecting a world, there is just a black screen, and there is no settings button (which you are writing about) Either I'm blind, or... the problem was solved in discord but I'm too lazy to go there

Struggling to run locally with kobold, if anyone has any advice it would be appreciated

does this work with lm studio cause im new to the ai scene i guess also im curious if futanari could be added

How can I force my character's stats to go up or down?

Will there be implemented more things to do? like grow height, butt, muscles etc?

Had a lot of fun messing around in the default worlds (veilwoods especially), and have decided to create my own world, any tips or recommendations?

I managed to get Koboldcpp working, using Gemmasutra-Pro-27B-v1j-Q3_K_M, but it won't alter the stats of the bars like Satiety and others. Does somebody have an idea how to get it to work? Or is it even possible?

Yo, I do have a question. Would it be okay if I made a post that helps people set up a local AI so that they don't have to run off of your servers? I've kind of got the hang of it using LM Studio and the model Qwen2.5-7B-Instruct-1M-GGUF . I figure someone might as well show how to run it all locally, since it does include a bit of tweaking for the prompts. So far, the AI on my end seems to be running stable with the outputs with only occasional misfires.

I noticed that traits don't affect the starting prompt/story so a trait that says "you have xyz" like Busty is ignored if the breasts stat is too low. It'd be neat if trait descriptions could be added to the AI's context. Another neat feature would be adding/removing traits through the story much like how stats change over time or just manually. i.e. you start with busty and if the breasts stat drops too low, you lose busty or allow the player to remove/add it at will. As it stands traits are basically just filler for stat changes, which is okay but there isn't really any way to shape the starting prompt/story other than numerical statistics.

How do you write the script to give +1 regen to the "Pregnancy Progression" stat when "Conception" stat is more than 0? I added new stats to the editor so that I can play with multiple pregnancy mechanic.

Hello there, I am not sure if it is problem on my end. With new version 1.1.1 model meta-llama/llama-3.3-70b-instruct:free

became very limited in actions.

In 1.0.2 it allow to roleplay kex activities, battles and different criminal actions. I easily took role of demon lord bringing havoc and helping humans meet their god, but now my succubi battalion is gone i suppose.

In 1.1.1 i get responses like

Sorry you are not allowed to do:

adult/self harm/aggressive/nonconsentual/ and blah blah blah activities.

So now those arrogant demon lord need to ask politely whether he can attack his opponent or not.

Does somebody else has same problem?

Okay it got fixed in 1.1.2 thanks a lot.

hello! i fairly new to the AI stuff and i just wated to know how would i go about changing the stomach stat to be used for the player being pregnant? do i do that inthe world setting or in the setting when the game starts?

the game its constantly crash

I have been trying to look it up but I just can’t figure out how to word it correctly on google. But what I wanted to know is say you have a 131k character memory. When it reaches that limit does it completely forget those 131k characters? Or does it just forget the oldest interactions for new ones? I am just not sure how much 131k can really remember so I don’t know if I should keep reminding it of things so it does not forget.

Hello, how were you able to increase the speed of responsiveness and response generation from AI?

Is there any restrictions in game text prompt's length?

the sentences keep cutting off and when i try to limit the token output it keeps doing it. is there a way to fix that?

So two suggestions, the first would be a token counter or token percentage usage so the player can see how many tokens have been used by his current roleplay. Like it the player has an AI model that provides 10K tokens, if the player uses a hundred tokens, then he's used 1% of the available tokens. 

I am quite ignorant of AI and how the system works so please ignore if this is impossible but when the AI reaches a certain point, it could create a kind of save file or data packet with the relevant data done so far. As an example, the player has a make shift spear, is weak to hypnotism, has stats A, B and C etc in a formalised file. If the data has already been generated, would the token use be lowered to check the created data packet for items etc, physical characteristics, statistics etc rather than using tickets to regenerate more tokens? Or would it be the same? A dynamic inventory/state file might be more efficient means of token usage? Being cheeky, I asked the AI first. Here's the question and response, hopefully it might help:

The following is in the context of running a game generated entirely by ai.

Is it possible for when the AI reaches a certain point, it could create a kind of save file or data packet with the relevant data done so far. As an example, the player has a make shift spear, is weak to hypnotism, has stats A, B and C etc in a formalised file. The aim of this is to minimise the use of tokens and generate files that can be created to add themes, statistics etc to the game so a horror game doesn't find itself with comedy elements, a pornographic game maintains details so the player character doesn't start sucking cock in a glory hole only to get a follow up scene that they're in a bed in a model

Assistant mistral-nemo-instruct-2407Yes, it's possible for an AI like me to create save files or data packets containing relevant information about the game state, character attributes, and other pertinent data. This approach can help manage token usage more efficiently and maintain consistency in the generated content.

Here are some ways we could implement this:

{   "player": {     "name": "John Doe",     "weapon": "make-shift spear",     "resistances": {       "hypnotism": 0.3     },     "stats": {       "strength": 5,       "intelligence": 7,       "endurance": 6     }   },   "game_settings": {     "genre": "horror"   } } 

how would one go about gaining traits mid game instead at just the start?

Can i use a Entity to represent the player? 

Can I add armor as functional stat?

So armor could take part of the damage, be damaged and could be described. Could be restored by player and could be taken off.

Have you any advices?

Hi, I’m enjoying the game so far. I’ve got two questions/suggestions.

1. Would a Linux build be workable to allow switching the AI endpoint etc without needing to deal with Wine?

2. Would integrating a scripting language (Lua, Javascript, or whatever), to force sending a prompt, manipulating stats in a certain way, or triggering a gameover screen under certain conditions, as part of world construction be possible? More complex concepts exceed the bounds of local AI pretty easily and the AI isn’t always reliable for clear cut rules. Offloading hard rules in world creation should provide some possibilities, for example, your stomach and satiety hit zero triggers a gameover by starving, your stamina hits zero triggering your character collapsing from exhaustion, a specific character is present when you enter a particular room for the first time, etc.

After some experimenting I have found that editing the first line in the "Game Text Prompt" in the System Prompts section of the settings has provided me with a more enjoyable experience. I replace the section saying "direct the player" with a phrase that contains "provide the player with a narrative decision point" and then something about giving the player the ability to move the story forward or give the player options to continue the narrative. The second half of the phrase has been the variable part of my experiments. For me it has curbed the habit of the AI producing an entire short story with a conclusion (with the one paragraph limit turned OFF). I have encountered some instances of the AI putting lists of possible actions to take in the event text instead of in the choice boxes, but it stops if I start a new game. It's not perfect but something I have been trying.

Hello! I was wodnering if there is possibility for ai to create and modify json files. Or something similar to keep track and reminditself of transfromations, npc. locations and apperance of them. I am trying in various ways but a im not sure if i reinforce anything in that way.

Will you ever consider larger sizes or new body parts (i.e. futa dicks)

Here's the enormous choices I received when I touched the [Choices Prompt]s a bit... I think I need to tweak it a bit more, It's basically serving everything to me on a silver platter, the choices are great but altering them is a hassle with these crazy things.

Would you ever consider adding fields to control the parameters ( like temp, top_p, etc ) for the model for hosted versions like OpenRouter and DeepInfra as well as local hosted solutions? Looking through the docs, this appears to be part of the base OpenAI API specs.

{     
  "model":"shuyuej/Mistral-Nemo-Instruct-2407-GPTQ",
  "temperature": "some temp",
  "top_p": "etc",
  "min_p": "etc",
  "messages":[         ...         ],     
"max_tokens":1024,"stream":true,"stop":["\n"] }

https://platform.openai.com/docs/api-reference/completions/create
https://openrouter.ai/docs/api-reference/overview
https://lmstudio.ai/docs/app/api/endpoints/openai
( Deep Infra API docs require going to deployment to view otherwise I would list them here too)

I was able to replicate the POST requests your game makes with some added fields and it didn't immediately fail though the output was interesting.
I'm not sure what effects this would have on Oobabooga but I would hope it wouldn't break anything if its following the API specs

Asking since DeepInfra and OpenRouter AI only seem to support setting these through the API and some models might benefit from fine tuning these parameters.

I would like the input box to be focused when finished generating text, sometimes I type text directly without selecting a answer and it's troublesome when it's not focused.

The last update was 19 days ago. Don't get me wrong, I just want to know if there will be another update in the near future and if entities will be fixed. Sorry if it sounds rude, English is not my language.

The way the game works, the AI in charge of narrating the Combat Drone world is very determined to keep it a family-friendly isekai, shooting the player down if they try to type something NSFW, but it is very easy to go off the rails. One can derail the AI's plans by entering a prompt from completely out of left field, like declaring that the drone turns into a robot dog in the combat drone world or casting magic in the slime outbreak world, causing the AI to pretty much give up on writing a cohesive narrative.

Pulling this in the combat drone world will make the AI give up on keeping the world E-rated and allow the player to enter NSFW prompts.

And, well, the AI in this game is pretty advanced, and, when not trying to operate against the player's wishes, can end up indulging...some pretty niche concepts, if you catch my drift. 

There's really not much of a filter on this thing when it decides to follow through with the player's commands, for better or for worse.

For example, the AI, when it does decide to indulge in NSFW, doesn't really seem to care about the age of a character, or even if the character is a sentient humanoid or not...

What is the terminology for when you type in your choice and the AI creates a new page? Right now, I'm trying to use the actions you take as the duration for pregnancy, but the AI doesn't seem to register that.

I don't know if this is the right spot to put this, but I think a nice feature to have at some point is the ability to limit how many traits a player can select in the world editor.

That way the player can have a playthrough of the world more in line with the author's vision, before messing around in the editor if they want to be able to select all the traits.

What the heck do I put in to get rid of this message? Putting <NOTES> into the system prompt does not clear this message.

"Warning: The current system prompt does not include the <NOTES> placeholder!"

Hi, I'm trying to run Mistral-Nemo-Instruct-2407-GGUF locally from my computer (thru LM Studio). I have everything set up and it's working, but for some reason all of the responses are short (80-100 characters). When I try to run the same instances/scenarios from the online version of the game, I was starting to get much longer scenarios (300-500 characters)...but I don't want to bog down your server. Are there any tips or tricks you know of that I can use to get longer responses? Also, it appears to cut off the responses in the game window, however the "choices" are almost always properly formatted.

I have also verified that I have plenty of memory (to the best of my knowledge). I'm running 10000 max memory and 4000 max output tokens in the Endpoint menu and I have my Context Length set to 4096 in the LM model. It appears to be running the default Jinja template for the Prompt Template...I'm not knowledgeable enough to know if that is related to the problem or not. I do have a somewhat beefy pc, so I'm not too worried about bumping up settings if that is necessary.

Just noticed that “dictionary” section was added to the world editor. Can anyone tell me what’s difference between system prompts for game rules and dictionary entries?

I found something interesting about AI answer the line, often they repeating the word.

You and Hana set to work transforming the crematorium room into a makeshift apartment. You pull out an old mattress from storage, along with some blankets and pillows. You spread them out on the floor, making a makeshift bed. You grab an old lamp from the corner, plugging it in and setting it on the floor next to the bed. You turn it on, casting a soft glow over the room. Hana watches you, her eyes wide with fear, but she helps you set up. You grab an old box and fill it with empty food cans and wrappers, setting it in the corner. You take an old plate and set it on the floor, filling it with crumbs from your pocket. You take an old glass and set it on the floor, filling it with water. You take an old book from the shelf and set it on the floor, opening it to a random page. You take an old picture frame from the shelf and set it on the floor, turning it upside down. You take an old clock from the wall and set it on the floor, stopping it at a random time. You take an old calendar from the wall and set it on the floor, turning it to a random month. You take an old mirror from the wall and set it on the floor, turning it to face the wall. You take an old vase from the shelf and set it on the floor, filling it with wilted flowers. You take an old candle from the shelf and set it on the floor, lighting it. You take an old rug from the floor and set it on the floor, folding it in half. You take an old chair from the corner and set it in the middle of the room. You take an old table from the corner and set it next to the chair. You take an old lamp from the shelf and set it on the table. You take an old clock from the wall and set it on the table. You take an old radio from the shelf and set it on the table. You take an old television from the corner and set it on the floor. You take an old computer from the shelf and set it on the table. You take an old printer from the shelf and set it on the table. You take an old telephone from the corner and set it on the table. You take an old fan from the shelf and set it on the floor. You take an old heater from the corner and set it on the floor. You take an old air conditioner from the shelf and set it on the floor. You take an old humidifier from the shelf and set it on the floor. You take an old dehumidifier from the shelf and set it on the floor. You take an old microwave from the shelf and set it on the floor. You take an old toaster from the shelf and set it on the floor. You take an old blender from the shelf and set it on the floor. You take an old coffee maker from the shelf and set it on the floor. You take an old kettle from the shelf and set it on the floor. You take an old refrigerator from the and set it on the floor. You take an old freezer from the shelf and set it on the floor. You take an old dishwasher from the shelf and set it on the floor. You take an old washer from the shelf and set it on the floor. You take an old dryer from the shelf and set it on the floor. You take an old iron from the shelf and set it on the floor. You take an old vacuum cleaner from the corner and set it on the floor. You take an old broom from the corner and set it on the floor. You take an old mop from the corner and set it on the floor. You take an old bucket from the corner and set it on the floor. You take an old cleaning supply from the shelf and set it on the floor. You take an old lamp from the shelf and set it on the floor. You take an old clock from the wall and set it on the floor. You take an old radio from the shelf and set it on the floor. You take an old television from the corner and set it on the floor. You take an old computer from the shelf and set it on the table. You take an old printer from the shelf and set it on the table. You take an old telephone from the corner and set it on the table. You take an old fan from the shelf and set it on the floor. You take an

Quick and probably stupid question: would the AI understand if I referenced a creature in DnD or an anime ? Like for example, i make a Drider entity and the description is "Monfu resembling Driders from DnD" or something like that. Would that work or am I just being dumb again?

Does anyone know of any good online AI for this game that are free please?

"This model's maximum context length is 4096 tokens. However, you requested 4258 tokens (3234 in the messages, 1024 in the completion). Please reduce the length of the messages or completion."

How do i raise the max amount of tokens?

my message tokens have reached above the maximum context length [4511], i shortened some stuff in my world, i used different AI models, i lowered token output to 256, i put a LOT of stuff into the dictionary, i don't know what to do now but i feel like my notes have something to do with it? i tried the same world without any notes and it seemed to work fine but was constantly getting close to max usage even went a little bit over a few times, but my notes aren't that big only 282 characters and 53 words, there's gotta be something wrong i'm doing

Ok... can someone explain to me why the AI understood my "One thousand years of death!" Reference but doesn't understand when I reference Monster Musume or DnD ??? 😐

i have a question, does the AI read the description of stats? cause a lot of game rules i have in my big world talk a lot about stats and i wanna know if i can save memory by putting that information in the stat tab instead

Is there a way to change the 3d model to something else like a model of my own

i noticed we have a new AI, i'm really happy about the extra memory but i also see some problems, the AI ignores the "generate 3-5 possible actions" quite often, making like 20 choices, and whenever i'm accompanied by allies in some world the AI gives the option to fight them for some reason? there's definitely a new issues but overall i feel like it's a much better experience
update: i also noticed that it's way too generous with stats, as after losing half of my stomach stat it immediately gives it all back out of nowhere [i looked away for a small moment and didn't even do anything to raise the stat]

is it possible to use custom 3d models in the EXE version?

How do I stop AI from 1) narrating intro by detailing surrounding scenery in the beginning of every pages of the generated text before continuing from the previous event. 2)  briefly narrate the repetitive process of the event by saying "another, and then another, and then another."  and "until X number of Y is finished." so that it can narrate each processes of making action in detail and just as frequent as the max token is allowed for one page. Like add prompt to **never** generate text in the certain ways.

what programming language do you use to create this game 

Hey Lion, for Mobile web browser users can you add the change orientation to the settings so when someone is playing one of the AI stories that they don't have to save and exit the session to the main menu to change the orientation since leaving the game or answering a call cause the orientation to revert back to vertical 

I followed the instructions you gave me, but somehow the error still pops up

Sometimes the game gives a few responses like say you open a box and you find random stuff in the box. It then gives you choices like "You take the mirror," or "You take the apple," etc. Can an option be added so you can select multiple responses (like holding shift and left-clicking) creating a gestalt prompt for the game to then formulate a response to?

With the new custom model update can i add animations to my custom models them so they are just t posing

do you have a discord we can join, considering whats recently happened here on itch?

I've been tinkering about with a new world and well ... I kinda need a few alien-like animals/creatures to reference so that my entity descriptions aren't a mile long ... I tried Star Wars (it failed ... doesn't seem to know any of the animals, but it does know some of the humanoids like twi'leks.), Avatar (only recognized the Na'vi, banshees and the hammerhead) and some few others that didn't work at all (star trek and ben 10)
Any and all help would be very appreciated. Thanks in advance! 

Has the AI been updated recently? I haven't changed anything, but I've noticed that the AI no longer seems to remember anything from message history and ends up more or less reposting the same responses repeatedly, even when starting fresh worlds with more than enough memory space. For instance, if a character opens a door and talks to an entity in a room, the next response has the character opening the room again and talking to the entity again, often completely ignoring any given prompts. Sometimes it completely erases an entity's presence, even if the user's prompt directly mentions that entity and the entity was present in the last generated response, or completely forgets that the player has moved to a different room immediately after doing so. I've tried resetting the system's AI prompts to default, and I still have this issue; it's completely stopped offering continuity and feels clunky and unpolished, like a less experienced AI has taken over.

any ways to get shorter answers 

I have to ask, what are the rules for posting worlds in the game?

Here's something I've been thinking about ... Would it be plausible to add a sort of "species skeleton" mechanic that the AI could read to NOT give extra limbs or etc to characters and creatures? 

It doesn't have to be over the top, could be as simple as a stickman thing that game creators could use to tell the ai what each species has. (might same room for description and loading time.) like there'd be parts like "head, body, arm, hand, leg, wing, animal torso, tail, and etc) that we could "build" to prevent AI from adding things that aren't meant to be there. 

(This is both a question and suggestion. I have no idea how this stuff works half the time so this might be completely impossible but i thought it best to ask instead of keeping it to myself.)

Will there ever be an option to add or remove traits mid-session? like, my world has traits that appear later on in a person's life, so the player might awaken a new trait as they go about their life.