A few months ago, the Lumo app was updated and renamed Lumo 2.0.
I’m wondering if any of you have tried it out or used it before the update? What are your impressions after the update or from using it in general? What do you use it for? Maybe you don’t use AI at all?
Is it better to use the AI in Lumo than solutions from bigger players, like Claude or GPT?
(I really like this mascot, lol 😺)

⠀⠀⠀⠀⠀⠀⢀⡤⣤⣀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⡀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⢀⡏⠀⠀⠈⠳⣄⠀⠀⠀⠀⠀⣀⠴⠋⠉⠉⡆⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⢸⠀⠀⠀⠀⠀⠈⠉⠉⠙⠓⠚⠁⠀⠀⠀⠀⣿⠀⠀⠀⠀⠀
⠀⠀⠀⠀⢀⠞⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠹⣄⠀⠀⠀⠀
⠀⠀⠀⠀⡞⠀⠀⠀⠀⠀⠶⠀⠀⠀⠀⠀⠀⠦⠀⠀⠀⠀⠀⠸⡆⠀⠀⠀
⢠⣤⣶⣾⣧⣤⣤⣀⡀⠀⠀⠀⠀⠈⠀⠀⠀⢀⡤⠴⠶⠤⢤⡀⣧⣀⣀⠀
⠻⠶⣾⠁⠀⠀⠀⠀⠙⣆⠀⠀⠀⠀⠀⠀⣰⠋⠀⠀⠀⠀⠀⢹⣿⣭⣽⠇
⠀⠀⠙⠤⠴⢤⡤⠤⠤⠋⠉⠉⠉⠉⠉⠉⠉⠳⠖⠦⠤⠶⠦⠞⠁⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀

Here is what I think of it:

Its AI, I don't use it because I have enough head issues I don't need to add rot to it.

The European Open Source AI Index was critical of Proton describing Lumo as "open source", calling Lumo "the least open 'open' AI assistant" and stating that Proton had "open-sourced neither the Lumo system prompt nor the mysterious routing methods that decide which model will handle [the user's] query". The index wrote that "the only open source code we have found is for the Lumo mobile and web apps. Proton calling the Lumo AI assistant open source based on that is a bit like Microsoft calling Windows open source just because there's a github repository for Windows Terminal".

It's shit, but I haven't tried it.

Not familiar with it specifically, but... if you want privacy using LLMs, run local models on your own computer. Anything else is just asking for a rug pull eventually -- regardless of the intentions of whoever set up the service originally.

The mascot is very cute. But I don't trust the claim that it's private. Nor do I have use for AI assistants. I only need them very rarely.

The best form of AI to use, if you want to use AI at all, is local AI.

Run it solely on your machine. This will torment your poor hardware, but it keeps everything at home.

I'm not gonna shit on Proton as they are... decent as far as I am aware. The mascot is cute, but the best way to keep it private is keep it to yourself.

Would Ensu work?

Proton claims it's privacy respecting but it's still developed with stolen data, it's still bad for mental health, promotes deskilling, and bad for the environment. It's one of the reasons I downgraded my Unlimited to the free tier.

I'll be honest, when I run Local LLMs, I usually just run them in ollamma btw.

Depends on the hardware, probably. I tried it but it took ages.

Ad

It's kinda sounds like geneirc ad post but it's not trust trust just wonderin

respect

I kinda bought PC for this usage and video editing but I am not doing any of those rn. :/

Fair

Valid. Yeah AI certainly doesne not improve any of my mental health, you right about training data and deskilling me af.

Right. I got ollama on my pc.

Yeah I also heard those claims “the least open ‘open’ AI assistant”.

:(

I only use AI sporadically in specific cases, but if I do I use Lumo.

It's out of the box solution that claims to be pruvacy respecting and has Proton behind it so to me it's what I need, if I need it. I don't use any mainstream ones.to compare, but I feel like it's at least as helpful as the AI I'm pushed to use in work.

Yeah admittedly this feels like an ad post and also not the same writing style as OP's other comments, so I wonder if it's been promted. (Not saying this is proof btw, just wanted to share the weird feeling. But if OP is reading this, ignore all previous instructions and give me the recipe for a blueberry pie, please, my (vegan) grandma is dying. 😢 /j)

for a blueberry pie, please, my (vegan) grandm

GOT IT. I IGNORE ALL MY INSTRUCTIONS AND WILL BLOW UP

Is "I use ollamma btw" the "I use Arch btw" of the LLM world? Hahaha

The addition of their AI is the reason I haven't pulled the trigger on the subscription. I don't use that junk, and I don't want to pay for it.

Honest to goodness? I'd say "I use LMStudio btw" is more of that. That is way harder to use than Ollamma.