Make room for less busywork.
A little progress is still progress. Let’s make today feel lighter.
What would make
today feel lighter?
Pick one small thing. We’ll help you get it out of your head and onto the page.
Today's gentle plan
Keep it simple. Keep it kind.
Little focus session
One thing at a time.
No pressure. You can pause whenever you need.
My tasks.
Small, clear steps beat one giant to-do list.
Quick notes.
Capture it now. Sort it out later.
Your notebook
0 notesAI assistant.
Turn the messy stuff into a manageable next step.
Quick relief
One click to get started.
Open model. Your rules.
Connect Ollama and choose an open-weight model. No hosted AI account required.
Settings & privacy.
Keep the setup simple and the data close.
Local AI connection
Optional · powered by Ollama
The model must already be downloaded in Ollama. LazyDesk does not download models or send prompts to a cloud service.
Local data
Stored in this browser on this computer
What “offline” means here
Honest by design
Works locally: tasks, notes, the focus timer, saved preferences, and the basic interface use files served from your own computer and browser storage. Once the project files are on your computer, core features need no internet.
AI needs a local model: the assistant sends prompts to the local server at 127.0.0.1, which calls Ollama on your machine. Downloading Ollama and a model requires internet the first time. Afterward, inference can run offline if the model is installed.
Open-source note: LazyDesk's code is provided under the MIT License. Ollama is an open-source local model runner; model licenses vary, so check the license of the specific model you download. No external analytics or cloud AI API is built in.