Proton

9436 readers

11 users here now

Empowering you to choose a better internet where privacy is the default. Protect yourself online with Proton Mail, Proton VPN, Proton Calendar, Proton Drive. Proton Pass and SimpleLogin.

Proton Mail is the world's largest secure email provider. Swiss, end-to-end encrypted, private, and free.

Proton VPN is the world’s only open-source, publicly audited, unlimited and free VPN. Swiss-based, no-ads, and no-logs.

Proton Calendar is the world's first end-to-end encrypted calendar that allows you to keep your life private.

Proton Drive is a free end-to-end encrypted cloud storage that allows you to securely backup and share your files. It's open source, publicly audited, and Swiss-based.

Proton Pass Proton Pass is a free and open-source password manager which brings a higher level of security with rigorous end-to-end encryption of all data (including usernames, URLs, notes, and more) and email alias support.

SimpleLogin lets you send and receive emails anonymously via easily-generated unique email aliases.

founded 3 years ago

MODERATORS

ProtonPrivacy@lemmy.world

alex_herrero@lemmy.world

Nelizea@lemmy.world

Introducing Lumo 2.0 (proton.me)

submitted 3 days ago by steam_lover@sh.itjust.works to c/protonprivacy@lemmy.world

5 comments fedilink hide all child comments

you are viewing a single comment's thread
view the rest of the comments

[–] SuspiciousCarrot78@aussie.zone 2 points 12 hours ago* (last edited 10 hours ago)

It could be Qwen 3.5 397B-A17B (MoE) or Qwen 3.5 27B dense- both score comparably on the Artificial Analysis Intelligence Index (AAII). Lumo should come clean about exactly which models are being served at which quants; this surely isn't a state secret. The context window is obviously capped at 128K. I'd like to know if we're being served at least Q4_K_M - ideally better.

Given that GLM 5.2 has a 1M-token context window and Qwen 3.5 supports up to 1M via scaling (with 262K native), being served 128K makes quant quality all the more important. I'd like some assurance it's not IQ2_XS or similar.

I'm trialling Lumo AI and have paid for a month to test it. So far... results are mixed, trending towards good.

Issues:

It defaults to an internal search tool over live search, which causes hallucinations from stale or incomplete content. This is significant.
Runs out of context mid-chat (despite compaction), necessitating a page refresh to clear.
The latest Android app (downloaded directly from F-Droid) is abysmally laggy (this should be priority 1)
Cannot download generated images on Android.
Can only search across 200 files from an attached Proton Drive folder; please increase this limit.
On that: make it easier to attach / detach / update files from within Lumo. Adding to Files is not immediately updated on Lumos end, resulting in hallucinations against contents.
Usage limits are opaque. It would be nice to have a rough sense of what "unlimited" means in practice (likely a ~5hr/day total message limit) so I can plan my work accordingly. I am aware of the image use limits (100/day for paid users) per announcement.

I'm trialling Lumo because its privacy stance is appealing (and I cannot run GLM 5.2 or 397B on-device), but the issues above are dampening my enthusiasm. They're all fixable. So please, Proton team, take this as critical (paid) user feedback. @ProtonPrivacy@lemmy.world