LLM_Eras.jpg
[Hide] (988.2KB, 6274x1479) Since the usual suspects seem to be constantly gatekeeping this while the typical corporate inbreds are trying to destroy everything it's time to have a thread and generate at least some interest and informative value on recent AI/LLM developments without any bullshit.
>Have VRAM (preferably a recent gpu)
>Anything under 8gb vram is pretty effin bad(but it's not fully over and you can still easily generate cool art though if that is your thing)
>If you are okay with slowness you can run off a decently beefy CPU and ram
LLM Agent? Do this.
>KoboldCCP
>a .gguf model from a site like
https://huggingface.co/
>rtfm
Hopefully you should have a pc that can run QWEN e.g. programming aid something like Qwen2.5-Coder
Art? Do this.
>download Comfyui and the generation model you want
>rtfm
That is all you need. You'll figure out the rest yourself it's more fun that way.
We should be figuring out ways to train or adapt our own models too because LLM models are becoming increasingly censored and braindead because the mainstream cuck tech fags keep pozzing training data into newest models now gradually polluting ALL models(because they have billions and we are poor).
Furthermore they hate and exploit open source and llm's anyway, since their dream is to run this in the dystopian zognigger cloud to be used against the population. Instead of decentralized local AI's helping us uplifting our knowledge and increasing productivity outside of their control.
You should also be saving some of these models, especially the historic ones, and keeping them in your digital library for safe keeping. This area moves fast and there's no telling what will happen.