I have avoided AI (LLMS) because it didn’t seem nearly useful enough to warrant supporting these tech assholes. Is there anyway to use it locally on a normal device (16 Gb ram, 8 year old Geforce graphics, etc?). Can I just slow it down and get it to work locally?
You can run very small models with Ollama or Llama.cpp. It really depends on the type of work you want to do, but these small models won’t do anything super complex. In my opinion the models don’t start getting really good until around the 30GB size like Qwen 3.8 27b. You could probably still use a small model for some chat, search, rag, maybe simple script writing.
You can get the TernaryQuench version of Qwen3.8:27b to run great on a GPU with just 16GB of VRAM. I started using it last week and it’s been amazing.
I have a 4060 Ti 16GB and was using the 14b version for a long while but that ternary conversion tech is just enough to squeeze the full 27b version into my card with plenty of room for context 👍
Is that by penkia on hugging face? (How safe is running random models?)
Note that you would need to use Llama.cpp for the Mix of Expert type models like the Qwen mentioned, Ollama doesn’t have the flexibility of commands.
Ollama lags behind the cutting edge of Llama.cpp but not for that long. It can do all that stuff. Maybe not well, and certainly much less configurable, but for casual use it’s fine and it’s much easier and more convenient until you start needing more control.
Not bashing Ollama, it’s great for getting started, and finding new models. I just had to switch over when I couldn’t set up the MoE model. The good thing was Llama can use Ollama blobs, so no loss of the older stuff.
You might’ve been running an outdated version of Ollama or something, or maybe just a model that wasn’t set up in a way Ollama liked. It sometimes has some compatibility problems with some particular models or the way they’re configured/implemented, it’s sort of stupid how it happily pulls models from huggingface but without actually configuring them correctly in many cases, but generally it fully supports MoE and all that stuff. Generally you can trust that the ollama-hosted models will pull and run correctly, but huggingface (where the better and more interesting models live) can be a crapshoot, it’s just the way it pulls and configures them though, there’s not actually broken.
Ollama can run MoE. It just decides for you where to place things. Also the Qwen model mentioned is a sense model.
That wasn’t my experience, but maybe I just didn’t run across the right instructions, or it was the particular model. Llama lets you adjust parameters far more. Two sides of the same coin, really.
Ollama is just llamacpp under the hood. Ollama is just an opinionated version that automatically sets things up that will work for most people. Llamacpp definitely has more of those “nerd knobs” that you can configure for better performance. The flipside is that it is far more complicated. Granted using another LLM to help you makes it not so bad.