Category:
Created:
Updated:
Space Shooter Game
my autoswarmcoder script i making a huge progress with creating the game Space Shooter. the app is utilizes chromadb (vector embeddings storage and all created context), websearch (duckduckgo), ollama (with cuda, currently i'm using two tesla m40s, are also a support for heating my crib :-), variable classifier (semi-liquid neural network) and all the coding LLMs (but mainly qwen2.5-coder 32b for tasks and codestral 22b for code-writing) and ChatGPT doing all the rest of the work- what is overwhelmingly impressive. 
all the components are opensource- i can achieve a score over 3000 
PS: it don't need a complex setup of an environment to play the game. SpaceShooter.pex can be started from linux terminal (if python3.10 is installed and all the images and sound files are in the same folder) with "SDL_AUDIODRIVER=dummy python3.10 SpaceShooter.pex"
HAPPY SPACESHOOTING!!!
0
0
0
0
0
0
0
Attachments
Info
Featured Posts
https://github.com/DuskoPre/AutoCoder/wiki
i'm not just testing LLMs but also creating a semi-liquid neural network (advanced LLM) with chromadb:
https://github.com/DuskoPre/liquid-neural-network-with-chromadb-cache
it seems there is a mayor upgrade for LLMs possible trough classifications (managing-system) of inputs to the LLM. the model is still static but trough the implementations of the classifier it can be made semi-liquid.
#dichipcoin
