Shipped: Ollama chat

Ollama Chat is a conversational AI chat client that uses Ollama to interact with local large language models (LLMs) entirely offline. Private, fast, no cloud.

Offline chats with local LLMs, multiple concurrent conversations
Thinking toggle (paw, thinkingOn/thinkingOff, respects model capabilities fallback)
Web search toggle (globe) — the model searches the web and reads pages by itself (tool calling), SSRF-protected
File uploads — up to 10 files per message (txt/md/docx as text, png/jpg/jpeg as images)
Markdown viewer — open local .md files in chat area, rendered safely
Rename / Delete / Download .md / Regenerate per chat
Prompt commands /file, /dir, /image, /url, /memory, /remember, /forget, /do, /?, /tasks, /task
Workspace memory (USER.md + MEMORY.md) injected into every model turn
Settings page (#/settings) — workspace folder and Gmail IMAP/SMTP settings with a connection test; the app password lives in <workspace>/.env only
Scheduled tasks (cron) — daily/weekly reports run by the in-app scheduler or Windows Task Scheduler, saved to workspace memory; managed on the Tasks page (#/tasks)
Model browser, download, monitor, selection, streaming
Show/hide thinking, Markdown/text view, export as Markdown
Light / dark themes, responsive layout
Local-only server — binds to 127.0.0.1 and is not reachable from the network