Others Multiic [v0.5.22] [Stvoler]
- Thread starter aleksandr unconditional
- Start date
status10
Member
- Jul 18, 2019
- 199
- 292
Yes it is, it's just not at all user friendly unless you know what you're doing. You gotta spend a few hours on the AI subject in order to set it up.
I kinda did and it kinda worked. Just never understood the image production and played without it as I couldn't make it work.
In other words, if really interested, get ready to invest time and effort. Being into fiddling with such things of course helps greatly. Some ppl just are like that.
SavageKing
Member
- Feb 10, 2018
- 395
- 169
can someone tell me why the choosing the female option that instantly translate to be amale instead of a female?
AbiOrionsson
New Member
- May 12, 2025
- 7
- 15
I do not know if I understand your question correctly, but I hope my answer fits with what you meant:
Sometimes the AI seems to default to the player being male. Normally it corrects itself when you use the back button and simply redo the choice.
that is exactly what i mean the AI should have a user setting check were it auto correct itself in future and thank you
AbiOrionsson for the help
does some one has this problem too when using local models Incorrect API key provided: undefined. You can find your API key at
You must be registered to see the links
. edit 1 when using the image generator its now gone but i don't know for how long i only use local stuff
Last edited:
A technical question, if anyone can/will help I would appreciate it. I am using the local version. I have been trying different models for stable diffusion.
I get this error for a number of models I have tried,both for gguf models and safetensor models. I look for basic models. So, I am wondering,is this because the stable diffusion model is older in Multiic? The models I am experimenting with are 2024 and 25 mostly, found at Civitai. I am very new to this still, so my question might not make sense but I am hoping it does and someone out there might help me. Thanks!
I get this error for a number of models I have tried,both for gguf models and safetensor models. I look for basic models. So, I am wondering,is this because the stable diffusion model is older in Multiic? The models I am experimenting with are 2024 and 25 mostly, found at Civitai. I am very new to this still, so my question might not make sense but I am hoping it does and someone out there might help me. Thanks!
xmodo99
Newbie
- Feb 20, 2020
- 27
- 18
I haven't got it working yet, but the recommended models are all for Illusion models. Are you downloading the right type?
does the crator lurk here can i request that he adds abilty to read the notepad files ? i can save the log of convo but the program has no way to return to it ? also to make the free avatar one without personality cus im sick and tired of overiding the personality. i just want to talk to the AI to do some dnd like rp. or if anyone can recomend me free AI that will not cencor me or bill me out of ass that i can upload my notepad logs and resume my story cus i really want to continue it but at the meoment if i try to post it in the 2,5k charcahter chunks (the promt size window) it would take 200 promts, and while i am willing to do this and spend the time, the AI brakes and get locked up if i try. i usually lurk and dont say shit but im at a wits end at this point
I played this game for a few days and found it very enjoyable, so I would like to suggest some improvements.
- More GGUF model names recommended by the developer and examples of the reasons for their recommendations are needed.
I think most people who download this game will not know much about AI. Since I believe that the diversity of the model's responses is the most important aspect of this game, I think the user experience will vary greatly depending on which model is used.
For someone like me, who is completely new to Koboldcpp, simply finding the model name and downloading it from Hugging Face is quite a lot of work, so I think it would be easier to simplify the process and allow users to preview samples of each model's responses before making a decision.
- Even though both the AI language and game language are set to English, all AI-recommended responses are output in Russian.
- I suspect that in Sandbox, the background changes based on user interactions by default, but it doesn't seem to be working properly.
- As an AMD graphics card user, I would appreciate some guidelines for optimization.
It took me about one to two days of extensive searching to find a relatively stable value. I understand that it may be difficult for the developer to set up an AMD VGA environment, but detailed guidelines are still necessary.
Additionally, it would be helpful to have some warning messages and examples. For example, “When your VRAM is insufficient, the model's responses may become shorter, less intelligent, or repetitive. In such cases, try adjusting the settings as follows.”
- Lastly, based on what I understood from GPT, the model can remember conversations longer based on context sizes. I'm wondering if you could add a extension for this. I think it's a necessary feature considering users' limited graphics card VRAM size. Of course, this might just be my ignorance, but I'm curious if such a feature linked to the model is impossible.
- More GGUF model names recommended by the developer and examples of the reasons for their recommendations are needed.
I think most people who download this game will not know much about AI. Since I believe that the diversity of the model's responses is the most important aspect of this game, I think the user experience will vary greatly depending on which model is used.
For someone like me, who is completely new to Koboldcpp, simply finding the model name and downloading it from Hugging Face is quite a lot of work, so I think it would be easier to simplify the process and allow users to preview samples of each model's responses before making a decision.
- Even though both the AI language and game language are set to English, all AI-recommended responses are output in Russian.
- I suspect that in Sandbox, the background changes based on user interactions by default, but it doesn't seem to be working properly.
- As an AMD graphics card user, I would appreciate some guidelines for optimization.
It took me about one to two days of extensive searching to find a relatively stable value. I understand that it may be difficult for the developer to set up an AMD VGA environment, but detailed guidelines are still necessary.
Additionally, it would be helpful to have some warning messages and examples. For example, “When your VRAM is insufficient, the model's responses may become shorter, less intelligent, or repetitive. In such cases, try adjusting the settings as follows.”
- Lastly, based on what I understood from GPT, the model can remember conversations longer based on context sizes. I'm wondering if you could add a extension for this. I think it's a necessary feature considering users' limited graphics card VRAM size. Of course, this might just be my ignorance, but I'm curious if such a feature linked to the model is impossible.
I'm having the same problem with a 3060 Ti, also, most models in civitai are safetensors, do i need to convert them to gguf or is there a site with more gguf? (hugginface doesnt help either)
I have been able to use Illustrious models, both safetensor and gguf. By disabling my antivirus (and disconnecting from internet physically because I am paranoid), I am able to run all Illustrious checkpoint models I have tried(checkpoint just means, as far as I can tell, a full model...as opposed to Lora etc which are more fine-tuning of a model you have). I cannot get Flux models to to work. I have decided for fun to try different kinds of models and see if anything but Illustrious works. If any other kind works, I will let you know.
Multiic-0.5.22
rpdl torrents are unaffiliated with F95Zone and the game developer.
Please note that we do not provide support for games.
For torrent-related issues use here, or join us on
You don't have permission to view the spoiler content.
Log in or register now.
Please note that we do not provide support for games.
For torrent-related issues use here, or join us on
You must be registered to see the links
!
You must be registered to see the links
,
You must be registered to see the links
. Downloading issues? Look here.