Memory reclamation really useful if you are using Ollama or similar in WSL2 - after awhile, Ollama will unload the model and WSL2 will give the memory back to the OS

Yeah I really wish ollama didn't do that. I tried several ways to make it keep the model in memory like an environment variable but I've so far failed.

Shoot sorry about that. There's a few ways to keep the model loaded in memory:

1. If using `ollama run`: `ollama run llama3 --keepalive -1`

2. If running ollama serve directly, use `OLLAMA_KEEP_ALIVE=-1` ollama serve

3. If using the api, there's a `keep_alive` parameter you can set to -1

Yeah I tried all those things. Especially the middle one. Because that's how I use it mostly. I added the environment variable to the systemd service. But it still removes it after 5 minutes. Very weird.

If only they would fix the memory leak and freeze on resume from hibernate that has been an issue for the last year at least...

I just HOPE I'm mistaken but until they support reading and writing ext4 fs WSL won't feel complete to me

Not sure what you mean by this - the root filesystem is ext4 for example.

WSL is still very broken on Windows on ARM for any cpu intensive workloads. With the push to popularize this platform, I really hope some attention finally gets paid to the various RCU stall bugs that have been reported and ignored.

As long as wsl caches file transfers to windows in RAM, it's just a toy.

Do they still pipe filesystem calls through a windows share? lol