MCP.so
Sign In

Deep-Co

@succlz123

About Deep-Co

A Chat Client for LLMs, written in Compose Multiplatform.

Basic information

Category

Other

Runtime

kotlin

Transports

stdio

Publisher

succlz123

Config

No standard config provided

This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.

Repository

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Deep-Co?

Deep-Co is a chat client for large language models (LLMs), built with Compose Multiplatform. It targets desktop platforms (Windows, macOS, Linux) and plans mobile support (Android, iOS). It supports API providers such as OpenRouter, Anthropic, Grok, OpenAI, DeepSeek, Coze, Dify, and Google Gemini, as well as any OpenAI‑compatible API or local models via LM Studio/Ollama.

How to use Deep-Co?

Download the latest release from GitHub or build from source using Gradle commands provided in the repository. After launching, configure your LLM API keys, manage prompts, and optionally set up MCP servers (requires uv and node on macOS, winget packages on Windows). You can start chatting, use chat history, and adapt SillyTavern characters.

Key features of Deep-Co

  • Desktop platform support (Windows, macOS, Linux)
  • Chat with streaming and complete responses, with history
  • Prompt management with user‑defined prompts
  • SillyTavern character adaptation (PNG and JSON)
  • MCP (Model Context Protocol) support
  • Text‑to‑speech via Edge API
  • i18n (Chinese/English), app color themes, dark/light themes

Use cases of Deep-Co

  • Chat with multiple LLM providers from a single desktop application
  • Roleplay with imported SillyTavern characters
  • Manage and reuse custom prompts for different conversations
  • Use MCP servers to extend LLM capabilities
  • Run on desktop or compile for Android (experimental) with Gradle

FAQ from Deep-Co

What LLM providers are supported?

Deep‑Co supports DeepSeek, Grok, and Google Gemini (implemented). Claude, OpenAI, and Ollama are planned but not yet available. You can also configure any OpenAI‑compatible API or use local models via LM Studio/Ollama.

Does Deep‑Co support MCP (Model Context Protocol)?

Yes, MCP support is implemented. You need to install uv and node (macOS) or the equivalent Windows packages using winget to run MCP servers.

Which platforms can I run Deep‑Co on?

Desktop platforms (Windows, macOS, Linux) are fully supported. Android and iOS are planned but not yet available.

How do I install Deep‑Co?

Download the latest release from the GitHub Releases page or build the desktop distribution using ./gradlew :desktop:packageDistributionForCurrentOS. Android builds require ./gradlew clean :androidApp:assembleRelease.

Are there any known limitations?

Mobile platform support, chat message export, chat translation server, Claude/OpenAI/Ollama LLM support, RAG, and an MCP server marketplace are not yet implemented.

Comments

More Other MCP servers