Download the Windows installer from the downloads page and run it. You can choose a custom install directory, and shortcuts are added to your Desktop and Start Menu. Launch LexureChat, pick a model from the top bar, and start chatting.
Click the model selector in the top bar and pick from the list of configured models. Your choice applies to the current conversation — starting a new chat lets you pick again, so different chats can use different models.
Token usage for the active chat is shown next to the model selector, so you can keep an eye on how much context you're using.
Type your message in the composer at the bottom of the window and press Enter (or click send) to submit it. Use "New Chat" in the sidebar at any time to start a fresh conversation — your previous chats stay listed in the sidebar so you can return to them later.
You can attach files to a message with the attach button in the composer toolbar if you need the model to look at a document or image.
Open the Memories panel from the sidebar to see what LexureChat has remembered about you across sessions. You can review, edit, or delete individual memories here — nothing is stored that you can't see or remove yourself.
Skills are reusable instructions the model can call on while chatting — for example, the built-in image generation skill. Open the Skills panel to see what's installed, or add your own custom skills.
Click the "Search" toggle in the composer before sending a message to let that message use live web results. Toggle it off again for messages that don't need current information.
Open the Permissions panel from the sidebar to control what tools the model is allowed to use — file access, shell commands, image generation, and web search are all gated here. Nothing runs on your machine until you've explicitly allowed it.
Files and images the model generates for you — including images made with the built-in image generator — are saved as artifacts. Open the Artifacts panel from the sidebar to browse, reopen, or export them at any time.