Lumo
366 results found
-
The Whole Cutesy Cat Thing Is Annoying
The whole cute cat thing is annoying and childish. I strongly dislike the "purring the engines to life" and "chasing digital mice" types of loading messages every time I open the app. It's not funny or cute - it's just obnoxious. Frankly I am not a cat person and don't find any of this theming to be entertaining. The suggestion here is really to not impose your idea of Lumo's personality on the user but rather let us use the custom instructions and our own prompts to guide what kind of personality we want from the tool.
8 votes -
RAG my proton drive. Make Lumo be able to chat with me about my files.
I bought into Lumo and upgraded my proton drive as I had the impression that I could query my files.
In fact, I can only upload some of them. Well, some, in fact only 10Mb before it complains. Then if I leave the chat and come back, it seems to have forgotten I uploaded a file to chat with.Here is my use case which is currently available in CoPilot.
1. Have an option for Lumo to chunk and ingest the files in my drive.
2. Allow me to query them like a proper RAG systemThanks.
8 votes -
Create doc and spreadsheet directly from chat
Add the ability to automatically open a chat window directly within a document and embed a table in Proton Sheets. This would streamline real‑time collaboration by allowing users to start a conversation or view a related spreadsheet without leaving the current file.
7 votes -
Implement Bias Mitigation for Chinese Open-Weight Models (Qwen, Kimi K2)
As noted in Lumo's Privacy Documentation, the system currently routes queries to several open-weight models, including Qwen and Kimi K2. While utilizing these models supports cost-efficiency and open-source diversity, research (e.g., NDSS 2026, "S1761") indicates they contain significant internal censorship and narrative engineering aligned with authoritarian state interests.
Proton’s core mission is to champion digital rights, encryption, and freedom against state surveillance. However, deploying models with embedded propaganda or censorship mechanisms creates a contradiction:
- Cognitive Risk: Users may receive subtly altered information or suppressed viewpoints on sensitive geopolitical topics (e.g., HK, Taiwan, human rights).
- Value Misalignment: This conflicts with Proton's…
4 votes -
Wider response area (smaller margins)
On a tablet or portrait monitor, the response should take up the full width - currently there is a lot of wasted space on each side, with the response in a narrow center block.
On larger screens, I still find it too cramped - especially with rich results such as tables. Tables are often very compressed and hard to read.
18 votes -
Add support for de-googled devices
Lumo is unsupported without Google Play Services. Can't start the application, blocked during boot.
Using LineageOS. Should work without Google.19 votes -
Copy/export AI reply the way it preserves font/formating/headers/quotations
Would it be handy if a button to copy the AI answer would be multi-functional to allow copying answer also in a
- markdown
- HTML
- BBCode ("Bulletin Board Code")
- readable text (currently pasted formated text with quotations is not so easily readable when pasted into a plaint text editor/viewer)17 votes -
Using better open-source/weight AI Models
Hi everyone,
I am writing this because I think Lumo has so much more potential. Imagine them using a 1 trillion parameter Chinese open-source/weight AI model like it would give them so much more flexibility in new features and accuracy of up-to-date data. They could even train an own AI model that they can then use like OpenAI uses ChatGPT as an multimodal AI model without having to rely on others creating open-source/weight AI models for them to use plus they could customize it perfectly for privacy, speed and accuracy. This own AI model could be based of some other…
11 votes -
To make Lumo more human
Subject: Proposal: A "Dual-Loop" Architecture for Next-Gen LLMs Inspired by Neuro-Mechanical Models
Dear Research and Engineering Team,
I am writing to share a conceptual proposal for a new architectural approach to Large Language Models (LLMs), inspired by the neuro-mechanical theories of Lee Kent Hempfling (specifically the distinction between Short-Term and Long-Term memory loops).
While current LLMs excel at pattern recognition and statistical prediction (effectively acting as massive Long-Term Memory systems), they lack a mechanism for inhibition, self-correction, and proactive intent. This proposal suggests a "Dual-Loop" architecture that could significantly reduce hallucinations, bias, and reactive behaviour.
The Core Concept: Dual-Loop Architecture…
2 votes -
Multiple personalization settings
Would it be possible to add different personalization settings and switch between them depending on how lumo is being used?
11 votes -
Allow switching models to better models for plus subscribers
Add a model switcher for plus subscribers that lets you choose between powerful models like gpt-oss 120b, deepseek v3.1, Mistral small and medium, kimi K2, Qwen, etc.
15 votes -
Latest highly efficient open source LLMs suggestions.
I would like to suggest some of the latest, cheap and very powerfull llms, so that you can replace current LLMs for better ones, to improve consistency and manage costs.
hf.co/Qwen/Qwen3-Next-80B-A3B-Instruct
hf.co/Qwen/Qwen3-Next-80B-A3B-Thinking
hf.co/openai/gpt-oss-120b
hf.co/zai-org/GLM-4.5Vplease consider using them in lumo instead of the current models
15 votes -
Live Translation Capabilities For Lumo
Lumo, as a private AI, is primed to be an excellent assistant for users to have in their daily lives. This includes the ability to accurately translate pages and other media for the user, without compromising their media data, metadata, and contributing to the data collection of other larger and gluttonous chatbots that have this capability, like Google Gemini and OpenAI's ChatGPT.
9 votes -
Emphasize that user data is never used to train AI models
From the Lumo "Getting started" webpage: "Information you enter is never used to train AI models." I think this should be emphasized more and expounded on, perhaps giving scenarios using both Lumo and other AI chatbots. (It looks like a no-brainer choice, to me, of which bot to use given that all the other ones (Maybe not DuckAssist?) exploit you and your data/thoughts.) While the privacy mantra of Proton is well-known and can be assumed, that's no reason to be subtle.
20 votes -
Pincode to open Lumo
It would be nice if the user is prompted to insert a pin code when opening the app, just like on Proton Pass.
9 votes -
Ease up on the censorship within Lumo
I am an AI enthusiast and engineer. I am not kidding when I say I have not been able to prompt Lumo ONE SINGLE TIME without it saying that it cannot help with whatever I am prompting it for.
If you EVER want to sell this to ANYONE (especially if you ever want to sell the API), you probably aught to take the route that Venice AI is taking and make it an uncensored (or less censored) model. I genuinely could mention something loosely interpret-able as criminal, and it wouldn't even bother to try and reason whether it is safe…
10 votes -
Lumocli (console terminal client)
The ability to run Lumo in a text-based user interface (TUI) in a console (terminal) client, in for example bash or cmd. With no dependencies to a graphical user interface (GUI).
With access to read and modify files, and the ability to run commands.
6 votes -
could calendar event emails populate the event title in the email subject line please?
existing feature = calendar event email notification
if calendar event emails populated the event title in the email subject line then the event would be identifiable without needing to open the email and find the event title there. This would afford a process efficiency gain.7 votes -
Option to change ghost mode chat back to normal chat to be saved
Sometimes I started a chat in ghost mode, but changed my mind and would like to save it. This is currently not possible, so a button to revert ghost mode chat to normal chat would be nice.
2 votes -
Add the ability to upload images from Photo Library on iOS
There is no way to upload the images in my iPhone photo library as part of my prompt. The only options I see when I click the “Attachments” button are to either upload a file or select a file from Proton Drive. Even if I click “Save to Files” on my photo in the photo library, I can’t upload it to Lumo because, Lumo doesn’t support the iOS photo format.
6 votes
- Don't see your idea?