I’m building Tinta, a journal app for Android where all the AI runs on the phone. It works in airplane mode once the models are downloaded.
You write or dictate an entry, and a small model (LFM2.5 1.2B) asks you a question to reflect further. It turns out 1.2B models are decent at asking questions, so I leaned into that. An embedding model finds older entries about the same thing and shows bits of them under the new one.
Most people using it are on budget phones running the models on the CPU, where the app provides good experience.
Great idea! I'd love this, but as an open source app. Out of interest (not a criticism), what made you decide to do this closed source, especially considering there doesn't seem to be a commercial interest involved (for now)?
Thanks for a good question!
I see this app as an MVP that needs user feedback to deliver exceptional value in a world where cloud AI is abundant and doesn't use 2GB+ of RAM to run. I am excited to see what shape it will take, and it feels too early to determine source availability.