ToZ-Wide AI Answering System

Monophtalmos · 2025-11-14T09:47:27+00:00

Hello everyone, As the title suggests, I want to create an AI system for the forums that uses RAG. But currently I don't have time to work on it. The AI will function similarly to ChatGPT and answer questions people ask. Its responses will be generated from all the threads written on the forum (a...

Jan 6, 2026#20

SigTyrsaid:post: 1178604
Don't forget about GPU, most import part for the LLMs. The more vRAM the better. However even without GPU it will be useful.
I think we should focus on scalability of the project, rather than impressing anyone with the hardware at early stages. We must be resourceful and utilize most of the things that community can get hands on. Many of us has older hardware laying around and giving it a new purpose is a good way to reduce e-waste a little bit.
We will at some point need people who are familiar with crypto/blockchain technologies and somebody on the front end.

I decided one of the major questions we had to solve is assessing the CPU, GPU and RAM of all devices, so that this can be communicated and recorded by other systems. I wrote this in python, I assumed that was intended. I normally work with java script, but in this case python is too dramatically superior not to use.

At first I created a script that used too many dependencies and collected way too much information about my device. I was able to see absurd things like my IP address and even that I had an inactive VPN at the time. The version I settled on is much simpler and generates a device ID number without collecting hardware serial numbers.

Conceptionally, the block I created is run on each device to measure total, available and later on consumed resources. Other systems can request a diagnostic report for each device whenever needed. It appears this is necessary for effective management of resources. People will likely want to record their own copy of these diagnostic reports that go into more detail than the block currently does, I know I definitely will be.

I'm going to work on another block now, likely for managing multiple VMs based on the collected data.

I'm willing to share the scripts I've already worked on, not sure how I should submit them tho. Obviously, people should look at what the script does before they run it, even if they have it analyzed by an AI.

So ya this was already a lot of fun, hopefully people are interested in putting actual work into this and don't just like talking about it.

SigTyrsaid:post: 1178604
Don't forget about GPU, most import part for the LLMs. The more vRAM the better. However even without GPU it will be useful.
I think we should focus on scalability of the project, rather than impressing anyone with the hardware at early stages. We must be resourceful and utilize most of the things that community can get hands on. Many of us has older hardware laying around and giving it a new purpose is a good way to reduce e-waste a little bit.
We will at some point need people who are familiar with crypto/blockchain technologies and somebody on the front end.

I decided one of the major questions we had to solve is assessing the CPU, GPU and RAM of all devices, so that this can be communicated and recorded by other systems. I wrote this in python, I assumed that was intended. I normally work with java script, but in this case python is too dramatically superior not to use.

At first I created a script that used too many dependencies and collected way too much information about my device. I was able to see absurd things like my IP address and even that I had an inactive VPN at the time. The version I settled on is much simpler and generates a device ID number without collecting hardware serial numbers.

Conceptionally, the block I created is run on each device to measure total, available and later on consumed resources. Other systems can request a diagnostic report for each device whenever needed. It appears this is necessary for effective management of resources. People will likely want to record their own copy of these diagnostic reports that go into more detail than the block currently does, I know I definitely will be.

I'm going to work on another block now, likely for managing multiple VMs based on the collected data.

I'm willing to share the scripts I've already worked on, not sure how I should submit them tho. Obviously, people should look at what the script does before they run it, even if they have it analyzed by an AI.

So ya this was already a lot of fun, hopefully people are interested in putting actual work into this and don't just like talking about it.

Jan 7, 2026#21

Vicky<3said:
I was able to see absurd things like my IP address and even that I had an inactive VPN at the time.

There is no absurdity in that. Your external IP can be seen by anyone. This is how TCP/IP actually works in order not to abuse traffic. You can check the following out if you like. https://whatismyipaddress.com/

As for AI systems, I would suggest going with a simpler path. If the goal is to help people find information faster, then creating some database with software to search would be a way simpler solution than just trying to feed an AI model with whatever stuff possible and allow it generating random nonsense which is definitely a risk in all AI models.

I can understand some people preferring "chatting" with AI to find stuff, but this can be achieved without any serious hardware at all if this is really needed.

In any case, I am really glad people try finding solutions in order to make sure The Temple of Zeus is better appreciated and accessed by outside people. The direction this temple is going right now seems absolutely correct. Thus, this really gives lots of hope. Myself, I must follow a pretty different path, so I can not give all my expertise here as this would take too much time from me and divert from the path Gods clearly indicated to me. But maybe one time I will be able to come here and direct all my power towards improvements of this community. I really want that, but I also know what needs to be done before that, so I am just following what I must follow. When time comes, I will also present the path I have followed as this will be necessary for those who might follow a similar path right now. Still, I can give some expertise from my area of knowledge as I am a professional Software Engineer with University education and massive amounts of practical experience. In ways that would help people to direct their effort towards better paths.

To summarize, try understanding AI better and not treat it as a "bullet proof solution". Always try to focus on actual goals and use AI as a tool. Basically, use the least effort to achieve the most. There is nothing bad to optimize your work in order to achieve what is best for the ToZ, for Humanity.

Vicky<3said:
I was able to see absurd things like my IP address and even that I had an inactive VPN at the time.

There is no absurdity in that. Your external IP can be seen by anyone. This is how TCP/IP actually works in order not to abuse traffic. You can check the following out if you like. https://whatismyipaddress.com/

As for AI systems, I would suggest going with a simpler path. If the goal is to help people find information faster, then creating some database with software to search would be a way simpler solution than just trying to feed an AI model with whatever stuff possible and allow it generating random nonsense which is definitely a risk in all AI models.

I can understand some people preferring "chatting" with AI to find stuff, but this can be achieved without any serious hardware at all if this is really needed.

In any case, I am really glad people try finding solutions in order to make sure The Temple of Zeus is better appreciated and accessed by outside people. The direction this temple is going right now seems absolutely correct. Thus, this really gives lots of hope. Myself, I must follow a pretty different path, so I can not give all my expertise here as this would take too much time from me and divert from the path Gods clearly indicated to me. But maybe one time I will be able to come here and direct all my power towards improvements of this community. I really want that, but I also know what needs to be done before that, so I am just following what I must follow. When time comes, I will also present the path I have followed as this will be necessary for those who might follow a similar path right now. Still, I can give some expertise from my area of knowledge as I am a professional Software Engineer with University education and massive amounts of practical experience. In ways that would help people to direct their effort towards better paths.

To summarize, try understanding AI better and not treat it as a "bullet proof solution". Always try to focus on actual goals and use AI as a tool. Basically, use the least effort to achieve the most. There is nothing bad to optimize your work in order to achieve what is best for the ToZ, for Humanity.

Jan 8, 2026#22

SigTyrsaid:post: 1178951
For now, I would recommend focusing on a self-hosted personal use prototype, for querying provided information.
As much as I am really pleased with this enthusiasm and let's do it attitude, we must keep our heads cold and don't get ahead of ourselves too much.
P2P networking part is something to worry about later, but must be taken very seriously in term of cyber security.
For now experiment with Docker. See if you can get PyTorch running in Docker with access to GPU. Also try Ollama in Docker as well. Expose it to another container with an actual Python code that is communication with Ollama instance. Ollama would be used for dealing with LLMs. (It will be the main component of the node in the future)

Yep, basically I'm making something that is self hosted but can be upscaled later.

In certain aspects you're definitely right. For now, I'm still learning but also working towards a foundation we can perhaps and hopefully use later. I have multiple blocks created already, and plan to make more. It's starting to become more clear what is necessary here for this to work.

Starting next week I start my next semester in college and won't work on this as often, but I don't plan to drop this entirely. I'll likely be putting a few hours on weekdays unlike currently where I've been putting maybe 10 or more hours daily. That isn't saying my 10 hours accomplishes what you might in 10 hours.

I'm familiar with what you're saying, devices not working as if cursed is something I hear often. I've definitely been cursed before, but it's been awhile since it has happened.

I won't pretend, in likely grand hubris, that I personally enact the will of the Gods, but I'm fine with entertaining the possibility you might be. I've been hearing that sort of thing a lot lately. Often grand claims with little to show for it. I don't hold any burden of verifying these claims, that rests on the shoulders of those making the claim. There's also no point in challenging it, and hardly any reason to accept the reality of it either. What matters practically is produced tangible results.

SigTyrsaid:post: 1178951
For now, I would recommend focusing on a self-hosted personal use prototype, for querying provided information.
As much as I am really pleased with this enthusiasm and let's do it attitude, we must keep our heads cold and don't get ahead of ourselves too much.
P2P networking part is something to worry about later, but must be taken very seriously in term of cyber security.
For now experiment with Docker. See if you can get PyTorch running in Docker with access to GPU. Also try Ollama in Docker as well. Expose it to another container with an actual Python code that is communication with Ollama instance. Ollama would be used for dealing with LLMs. (It will be the main component of the node in the future)

Yep, basically I'm making something that is self hosted but can be upscaled later.

In certain aspects you're definitely right. For now, I'm still learning but also working towards a foundation we can perhaps and hopefully use later. I have multiple blocks created already, and plan to make more. It's starting to become more clear what is necessary here for this to work.

Starting next week I start my next semester in college and won't work on this as often, but I don't plan to drop this entirely. I'll likely be putting a few hours on weekdays unlike currently where I've been putting maybe 10 or more hours daily. That isn't saying my 10 hours accomplishes what you might in 10 hours.

I'm familiar with what you're saying, devices not working as if cursed is something I hear often. I've definitely been cursed before, but it's been awhile since it has happened.

I won't pretend, in likely grand hubris, that I personally enact the will of the Gods, but I'm fine with entertaining the possibility you might be. I've been hearing that sort of thing a lot lately. Often grand claims with little to show for it. I don't hold any burden of verifying these claims, that rests on the shoulders of those making the claim. There's also no point in challenging it, and hardly any reason to accept the reality of it either. What matters practically is produced tangible results.

Jan 25, 2026#23

Vicky, just follow your heart and do whatever you feel right now. This is the only actual way to collect experience. And know this. You will always be supported in this place, even if you fail at some point. Failure here is not an option. Not because we reject failed stuff. But because we see that even so-called failures are ways to learn some lessons.

Vicky<3said:
I'm also not trying to create something perfect, I know I certainly can't accomplish that.

Perfection is a variable which changes its value. This results in the fact there is no way to create anything perfect.

Vicky<3said:
What you're proposing to do instead isn't ambitious enough to motivate me to work on it. It's not exciting to talk about, and thinking about it doesn't give me the unconquerable drive this does.

People with huge ambitions are always welcome here. Just make sure your ambitions align with reality.

Vicky<3said:
I'm quite confident I can build something that will at the very least benefit me, and therefore as a consequence my lived ones. I'm not trying to save the world. We're getting very near the point where it'll be too late for many now, I made my peace with that.

This world is to be saved. Just not today. It appears Humanity is going to get another world to start all over. But even this is going to happen in a couple of hundred years. But these are details. What is crucial is actually believing in humanity. Humanity even today has huge potential to become the most powerful entity in this Universe, even this reality. Myself, I reject the idea that Humanity is doomed. I will never accept this type of thinking. Maybe in previous life times I was not even human, but now I am. And I see the full scope of the human psyche. And this psyche based on core is the most beautiful structure this reality has ever seen. Our species has suffered so much and yet we still rejected the choice to lower our level of consciousness. Which is why I believe that there will come a time when Humanity is going to become the most powerful force in this Universe and even beyond.

Vicky, just follow your heart and do whatever you feel right now. This is the only actual way to collect experience. And know this. You will always be supported in this place, even if you fail at some point. Failure here is not an option. Not because we reject failed stuff. But because we see that even so-called failures are ways to learn some lessons.

Vicky<3said:
I'm also not trying to create something perfect, I know I certainly can't accomplish that.

Perfection is a variable which changes its value. This results in the fact there is no way to create anything perfect.

Vicky<3said:
What you're proposing to do instead isn't ambitious enough to motivate me to work on it. It's not exciting to talk about, and thinking about it doesn't give me the unconquerable drive this does.

People with huge ambitions are always welcome here. Just make sure your ambitions align with reality.

Vicky<3said:
I'm quite confident I can build something that will at the very least benefit me, and therefore as a consequence my lived ones. I'm not trying to save the world. We're getting very near the point where it'll be too late for many now, I made my peace with that.

This world is to be saved. Just not today. It appears Humanity is going to get another world to start all over. But even this is going to happen in a couple of hundred years. But these are details. What is crucial is actually believing in humanity. Humanity even today has huge potential to become the most powerful entity in this Universe, even this reality. Myself, I reject the idea that Humanity is doomed. I will never accept this type of thinking. Maybe in previous life times I was not even human, but now I am. And I see the full scope of the human psyche. And this psyche based on core is the most beautiful structure this reality has ever seen. Our species has suffered so much and yet we still rejected the choice to lower our level of consciousness. Which is why I believe that there will come a time when Humanity is going to become the most powerful force in this Universe and even beyond.