kelivo

A Flutter LLM Chat Client. Support Mobile & Desktop.

What is kelivo?

A Flutter LLM Chat Client. Support Mobile & Desktop.

Overview of kelivo

What it is

Kelivo is an open-source, cross-platform LLM chat client built with Flutter. It supports Android, iOS, macOS, Windows, and Linux. The application allows users to connect their own API keys or sign in with supported subscriptions to use multiple large language models in one app. Conversations, settings, and files are stored locally, and there is no built-in account system. Backups can be made to WebDAV or S3-compatible storage.

Key features

  • Models and Providers: Supports native protocols for OpenAI, Google Gemini, Vertex AI, and Anthropic Claude, plus any OpenAI-compatible endpoint. Includes built-in presets for OpenAI, Gemini, Claude, DeepSeek, OpenRouter, SiliconFlow, Alibaba Cloud, Zhipu AI, xAI, and ByteDance. Allows account sign-in for ChatGPT (Codex), Grok, or Kimi Code.
  • Agent and Workspaces: Bind a conversation to a workspace to give the model tools such as shell, read_file, write_file, edit_file, and more. On mobile, runs a Linux sandbox (Ubuntu, Debian, or Alpine via PRoot on Android; Alpine via iSH on iOS). On desktop, tools run in the system shell.
  • Assistants, Memory and Context: Each assistant has its own model, system prompt, tools, MCP servers, skills, and default workspace. Supports long-term memory with auto-organize.
  • MCP and Skills: Connect Model Context Protocol servers over Streamable HTTP, SSE, or STDIO. Import skills from Markdown, .md, .zip, or GitHub URLs.
  • Device Tools: Includes time, clipboard, calculator, text-to-speech on all platforms; calendar and location on Android and iOS; screen time on Android; weather, reminders, and Apple Health data on iOS.
  • Scheduled Tasks: Run a prompt with a chosen assistant on a schedule (available on Android and desktop).

Use cases

Kelivo is suitable for users who want to interact with multiple LLMs from a single interface, manage conversations with file and code execution capabilities, and keep data local. It can be used for general chat, coding assistance, file editing, and automated scheduled tasks. The workspace feature enables the model to read, edit files, and run commands in a sandboxed or native environment.

Getting started

  1. Add a model provider: Open Settings → Providers, enter an API key for a preset provider or add your own endpoint, then fetch its model list. To use a ChatGPT, Grok or Kimi Code account instead, sign in from the Accounts tab under Add Provider.2. Start a conversation: Select a model from the input bar. Turn on web search, MCP servers, tools or reasoning for the current chat from the same bar.3. Let the model work with files (optional): Open Settings → Workspace & environment and create a workspace, then bind it to a conversation from the chat. On Android and iOS, install the Linux environment on the same page first; the iOS environment is bundled with the app.

License and pricing

Kelivo is open-source under the AGPL-3.0 license.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/chevey339kelivo.svg)](https://yunodb.com/tools/chevey339kelivo?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/chevey339kelivo?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/chevey339kelivo.svg" alt="Featured on YunoDB" height="20"></a>