@GamerBolt03
on
AEVA
12h 1m 48s logged
AEVA Devlog #7
Same progress as last time, the model is currently in training. I had some trouble back home and wasn’t able to train a model since my computer kept shutting down however the training will be done soon. Didn’t have an image to share lol.
9h 35m 9s logged
Air-Mouse Devlog #3
I haven’t really progressed much since before I’m currently in the middle of creating the UI since the base model and program work just fine. Also working on the icon for the app.
@GamerBolt03
on
Assignment Tracker
9h 12m 33s logged
Assignment Tracker Devlog #2
This project is practically completed I’m looking for any new features I should add or bugs, but I’ve turned it into a proper web app and can be downloaded as an executable. Here’s a simple icon I designed for the app.
@GamerBolt03
on
AEVA
16h 23m 47s logged
AEVA Devlog #6
The architecture is now very strong and is more than fast enough to reach the goal in 3 hours however I’m having issues creating the perfect dataset since the model becomes to focused on the roleplay datasets. I’m going to try using some online datasets as is for this to see if that helps.
16h 45m 41s logged
FunctionAI Devlog #4
Currently still working on improving the architecture and making it fast enough to learn object detection in at least under 30 mins.
Also thinking of working on the UI in the mean time since the project is almost complete
5h 5m 1s logged
WindowsGX Devlog #2
Working on fixing the UI since there are many small issues like scrolling and the covers being cut out as well as the meta data just being bad
@GamerBolt03
on
AEVA
4h 23m 47s logged
AEVA Devlog #5
Running into the same issue as last time but this time I think I have a better fix and will be distilling some model. in the meantime, I created a simple app icon for when I release it and will be working on the UI
13h 20m 46s logged
FunctionAI Devlog #3
After completely re-coding the entire project from scratch the new model is able to reach an accuracy of 30% as of now with 15mins of training on 1080p data and run at a framerate of 1000fps, which is not bad for having re-coded the entire project but I need to make it much better.
Also, I created a simple logo, although I couldn’t really get a good idea on what I should make
@GamerBolt03
on
AEVA
8h 33m 32s logged
AEVA Devlog #4
Training failed again, this is getting tiring but there’s no other real way to test the architecture. This time the training failed due to the dataset and how it was distributed. I was actually only giving about 0.13 tokens per parameter which was leading to it overfitting too. now I’m in the process of making the architecture more memory efficient and faster at processing data, as well as increase dataset size.
I also finally started working on creating a github repo as well as an explanation on how it works
@GamerBolt03
on
AEVA
3h 38m 25s logged
Well bad news, the AI infact did find a loophole, although after a lot of fixing as well as creating a better dataset, I’ve started training once again which will complete in another 3 hours, until then I’ll be working on another one of my projects. This is likely going to happen many times since I did decide to start doing such a hard project, I just wish I can finish the project in time. Also, for now I’m going to use this CLI Logo.
@GamerBolt03
on
Assignment Tracker
2h 4m 37s logged
This is a project I created because I always had trouble keeping track of assignments or forgetting them all together thanks to my goldfish like memory. That’s why I’m creating this project to help keep track of any assignments sent to me on my school or personal Gmail at the same time. This application allows me to keep track of any and all assignments using a simple UI interface. I’m actually mostly done with it I’m just doing more testing and adding any features I may want.
13h 26m 56s logged
Well after a lot of tests and coding (approx. 13 hours) I decided to switch to a neural based AI instead of function based using my custom CTF architecture. Ive been training the model on the COCO dataset with 80 classes for object detection. Currently in under 5 mins of training on an RTX5090 the model reaches 71% accuracy with estimated 1000 FPS on the Core Ultra 9 255k CPU which is insanely fast compared to mobile net and other datasets. I believe with an increase time limit the model can reach around 95% accuracy however I would like to keep the training time as is at 5 mins just to prove the architecture works insanely well too. Although might have to change the name due to switching to CTF.
(I really like sushi, dogs and video games)
@GamerBolt03
on
AEVA
15h 47m 54s logged
After another 16 hours of working on the project, The AI has finally reached a CE of 1.0!!!!! actually it reached a CE of 0.9 to be exact in only 3 hours of training which is just spectacular. Of course, the model still has to be thoroughly tested to see if it’s actually conversational since the model could have found a loophole to get a lower CE than it deserves. After some further testing ill post again within the next 7 hours.
5h 4m 27s logged
This is a project I started creating to support AEVA which is another project based around a new AI architecture however I decided that this deserved its own project since after all it may have no use in AEVA
Currently I attempted to only use function approximation along with grounding using linear regression however it only managed to achieve an accuracy of 10% which has led me to think of new ideas using a transformer replacement I developed for AEVA known as CTF as the star choice. However, I would still like to work further on the neuron less architecture a little further.
8h 0m logged
Working on creating a program to turn windows into a console like experience, the original reason I’m creating this program is because other programs like Steam Big Picture and Playnite were always difficult to navigate, had bad performance and themes or just weren’t seamless. I wanted to create this program for small old PC’s that can be turned int a gaming console without Linux to still have the full windows application support. With the start of this new event I decided to finally start working on it and ship it for some rewards. Currently I’ve partly designed the UI (User Interface) and have created functions to search the computer for games and apps and sort them allowing access to games downloaded from any provider.
1h 0m 57s logged
I just finished testing the program in many environments and its working perfectly which is great news however gestures and keyboard mode isn’t working at all which needs to be worked on but high chance main program will be done by today however ill still need to create a UI and built it into a proper .exe executable.
6h 5m 45s logged
WindowsX is a project I’m creating to solve the issues I have with older laptops, which often have poor performance and battery life (I use a 2019 MacBook pro 16 inch running bootcamp windows which disables the igpu so I’ve been dealing with these problems for a while with a previous battery life of around 30mins to 1hr that I even take to school). The goal of the project is to greatly improve battery life by dynamically switching system settings based on the current tasks and usage using a custom AI. It also removes unnecessary background processes to save RAM. My target is to save around 1 GB of RAM and increase battery life by up to 2 hours. However, I’m currently facing some issues testing the program. All of the functions work, but it’s difficult to properly test them because I don’t have the necessary hardware.
8h 56m 32s logged
I thought it would be cool if I could control my desktop from anywhere in my room with no problem instead of carrying a wireless keyboard and mouse to create a nice personal eco-system in my room, so I thought how I could actually do something like that and that’s how I got to this solution, although not practical without any other hardware it is cool and in school people might just think you’re doing black magic.
Air-Mouse is a program that lets me control my computer from anywhere using just my camera creating a virtual trackpad near you, currently the project is actually already almost finished most mouse related features work like left click using index pinch and right click using middle finger pinch and scrolling using the index and middle finger together. Currently I’m working on adding typing support by letting users toggle keyboard mode and draw letters in the air to type. I am also working on creating better custom models for gesture detection for better accuracy especially in dark areas.
@GamerBolt03
on
AEVA
13h 18m 2s logged
This is Project AEVA, an AI I’ve been trying to create completely from scratch on consumer hardware I own to compete with the current SOTA architectures. I originally started creating this AI because i thought that LLM’s although usefull and highly advance come with many issues esspecially on low-end hardware, halucinations, memory limit and the frozen weights after training. instead i want my AI to actually live with a human level life-span and continuously learn during that tmie slowly increasing its parameter size.
After months of designing and rewriting, AEVA (The name of the model I’m making) has officially reached its first training run. Instead of using Transformers, I’m building a completely new cognitive AI architecture designed to compete with frontier models while running on consumer hardware. The current architecture scales in O(n), is designed for massive context and persistent memory, and aims to support self-improvement and offline “sleep” learning in the future. Right now, the first model is being trained for 4 hours into training on my RTX 5090, and if everything goes well, it should become conversational soon. This is just the beginning for AEVA.
