Ollama
Reddit take
Not enough linked mention summaries yet; the source excerpts below are the evidence.
Pros
- Ollama for running local LLMs.
- 5- Obsidian or atomic calendar paired to Ollama
- I run local Ollama and run Cursor just fine.
- If you’re looking for lab things to do. Ollama and Open Web-UI are fun to play around with.
- From there, set up Ollama and OpenWebUI on the machine, and then set yourself up to access it remotely via a VPN app like Tailscale.
Cons
- Parent context: For instance I failed training to use a local ai with ollama. It works fine until I want to use it on a mobile device (like with the app Apollo for iOS) I was not able to access it trough a VPN on a mobile device. Now it...
- one hurdle I have is I host ollama, webui and my mcps on a truenas server via docker
- Is there a simple way to connect something like Ollama to reprocess or format notes locally?
- I recommend LM Studio over Ollama for a variety of reasons, but primarily because the former supports both GGUF (i.e. llama.cpp, like Ollama) and MLX, an inference format specifically optimized for better performance on Apple Silicon.
- Parent context: That's fair. I have Plex, and ollama running on it right now and I'll be honest and understand it's not a great reason, I just stress about breaking it. I fully acknowledge it is not set up the best. I went to update Olla...
Sentiment
17 positive (63%) · 9 mixed (33%) · 1 negative (4%)
Trust + time weighted score: +61% · raw score 59%
What redditors said (10 of 27)
Ollama for running local LLMs.
5- Obsidian or atomic calendar paired to Ollama
Parent context: For instance I failed training to use a local ai with ollama. It works fine until I want to use it on a mobile device (like with the app Apollo for iOS) I was not able to access it trough a VPN on a mobile device. Now it could be more a network setting issue. I tried to find the failure on my Unifi device, didn’t work. I feel like setting docker is difficult as Proxmox Reply: It's like setting up ollama for the first time. You follow documentation/ tutorials and eventually you will figure it out.
I run local Ollama and run Cursor just fine.
If you’re looking for lab things to do. Ollama and Open Web-UI are fun to play around with.
From there, set up Ollama and OpenWebUI on the machine, and then set yourself up to access it remotely via a VPN app like Tailscale.
one hurdle I have is I host ollama, webui and my mcps on a truenas server via docker
Run ollama for several friends, playing with stable diffusion, transcode like a champ, plus all the normal homelab stuff.
I setup an n8n workflow to automate a lot of this (yes could have used a cron job) tied to Ollama on my cluster.
I'm currently messing with ollama in a VM with m4000 passed through
Compare
Related
- Obsidian · related · 1,683 mentions
- Linux Mint · related · 728 mentions
- Proxmox · related · 543 mentions
- Fedora · related · 532 mentions
- Windows 11 · related · 510 mentions
- Windows · related · 413 mentions