I retired my 15 year old #Ubuntu LTS #laptop when I bought a 10 year old refurbished #Dell Optiplex 9020 with a #Nvidia #Geoforce 1050 Ti #GPU, so I can experiment with #GenAI, #MachineLearning and #LLM.

P. S. Yes I loaded #Linux on the Dell.

#FOSS #AI #ArtificialIntelligence #reduce #reuse #recycle #reduce_reuse_recycle #recycledtech

@chrisvitalos With only 4gb VRAM you're going to struggle doing "ai stuff". Stick to the smaller quantized modes like phi-2 or tinyllm and see how you go but you may need a gpu upgrade if you get hooked. I was running 7B 4bit (Q4) quantized models on a GTX 1080 8gb without issues but YMMV.

https://huggingface.co/TheBloke/TinyLlama-1.1B-1T-OpenOrca-GGUF/tree/main
https://huggingface.co/TheBloke/phi-2-GGUF/tree/main

TheBloke/TinyLlama-1.1B-1T-OpenOrca-GGUF at main

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

@chrisvitalos
Also cross-platfom free open source LM Studio makes it incredibly easy to download and try out many different models if you're not sure where to get started
https://lmstudio.ai/

Good luck!

LM Studio - Local AI on your computer

Run local AI models like gpt-oss, Llama, Gemma, Qwen, and DeepSeek privately on your computer.