SimpleLM
Context
A million tokens of context, kept on the device
Most on-device assistants lose the thread after a few thousand tokens. SimpleLM holds up to one million in a single conversation, and nothing is summarised away to save room.
What’s inside.
Private. Offline. Yours.What each one actually does.
Projects
Chat with your own PDFs and notes, offline
Add briefs, notes and PDF documents to a project and ask questions against them. Search runs locally on your iPhone or iPad, with no upload step.
Roles
Several perspectives inside one conversation
Apply a skeptic, an editor or a strategist and get a different reading of the same material, without starting a new chat or losing the context.
Vision
Read a screenshot or a document on device
Attach an image and ask about it. Text extraction uses Apple's Vision framework on the device, so a screenshot or scanned page never reaches a cloud service.
Private
Runs with no account and no connection
There is no account to create and no server to sign in to. Turn on airplane mode and the assistant keeps working, because generation happens on the device.