If you already didn’t know, you can run locally some small models with an entry level GPU.
For example i can run Llama 3 8B or Mistral 7B on a 1060 3GB with Ollama. It is about as bad as GPT-3 turbo, so overall mildly useful.
Although there is quite a bit of controversy of what is an “open source” model, most are only “open weight”
This isn’t true anymore, Intel dropped AVX512 since they moved to Big+Small cores design while AMD actually implemented it with Zen 4.