Its all local. Ollama is the application, deepseek and llama and qwen and whatever else are just model weights. The models arent executables, nor do the models ping external services or whatever. The models are safe. Ollama itself is meant for hosting models locally, and I dont believe it even has capability of doing anything besides run local models.
Where it gets more complicated is “agentic” assistants, that can read files or execute things at the terminal. The most advanced code assistance are doing this. But this is NOT a function of ollama or the model, its a function of the chat UI or code editor plugin that glues the model output together with a web search, filesystem, terminal session, etc.
So in short, ollama just runs models. Its all local and private, no worries.
I use borg backup. It, and another tool called restic, are meant for creating encrypted backups. Further, it can create backups regularly and only backup differences. This means you could take a daily backup without making new copies of your entire library. They also allow you to, as part of compressing and encrypting, make a backup to a remote machine over ssh. I think you should start with either of those.
One provider thats built for being a cloud backup is borgbase. It can be a location you backup a borg (or restic I think) repository. There are others that are made to be easily accessed with these backup tools.
Lastly, I’ll mention that borg handles making a backup, but doesn’t handle the scheduling. Borgmatic is another tool that, given a yml configuration file, will perform the borgbackup commands on a schedule with the defined arguments. You could also use something like systemd/cron to run a schedule.
Personally, I use borgbackup configured in NixOS (which makes the systemd units for making daily backups) and I back up to a different computer in my house and to borgbase. I have 3 copies, 1 cloud and 2 in my home.