ThinkInAIXYZ/deepchat
Deepchat
DeepChat is a desktop application that connects to multiple cloud and local AI model providers through a single unified chat interface, while also integrating external ACP-compatible coding and task agents as selectable models. It manages local AI agent sessions with project folders, permission modes, and resumable context for long-running tasks, and connects external tools and data sources via the Model Context Protocol using StreamableHTTP, SSE, or Stdio transports.
The application distinguishes itself by supporting remote desktop session control, binding messaging app channels to sessions for creating, switching, and managing conversations away from the desktop. It also loads task-specific instructions, reference files, and scripts from folders, ZIP files, or URLs to specialize agent behavior per conversation, and preserves structured session history with traceable context and tool calls for recovery and debugging.
DeepChat renders Markdown, code blocks, images, Mermaid diagrams, and artifacts within conversations, and supports retrying messages and forking conversations. It manages local Ollama models through a graphical interface without requiring command-line operations, and integrates web search APIs and browser simulation for retrieving external information during conversations. Sessions can be launched via deep links for seamless integration with other applications.
Features
- Selectable Model Integrations - Integrating external coding and task agents through a standardised agent communication protocol as selectable model entries.
- Multi-Provider Chat Interfaces - Connecting to multiple cloud and local AI model providers through a single unified chat interface without switching applications.
- Unified Provider Interfaces - Routes requests to cloud LLMs and local Ollama models through a single abstraction layer without switching applications.
- Cloud Model Connectivity - Routes requests to a wide range of cloud-hosted language models from a single interface, including OpenAI, Gemini, and Anthropic.
- Multi-Provider Conversation Managers - Connects cloud and local AI models in one interface so users can switch providers without leaving the conversation.
- ACP-Compatible Agent Runtimes - Runs ACP-compatible coding and task agents as selectable models with a native workspace UI for plans, tool calls, and terminal output.
- MCP Protocol Integrations - Connecting external tools and data sources via the Model Context Protocol using StreamableHTTP, SSE, or Stdio transports.
- Language Model Connectivity - Connects to cloud LLMs and local Ollama models through a unified interface without switching applications.
- Session-Based Agent Managers - Manages local AI agent sessions with project folders, permission modes, and resumable context for long-running tasks.
- MCP Tool Connectors - Connects external tools and data sources via the Model Context Protocol using StreamableHTTP, SSE, or Stdio transports.
- Local Agent Session Managers - Managing local AI agent sessions with project folders, permission modes, and resumable context for long-running tasks.
- Local Concurrent Session Managers - Manages concurrent local AI agent sessions with project folders, permission modes, and resumable context for long-running tasks.
- Chat Content Renderers - Displays Markdown, code blocks, images, Mermaid diagrams, and artifacts within conversations for diverse result presentation.
- Multi-Transport Connectors - Supports StreamableHTTP, SSE, and Stdio transports for Resources, Prompts, and Tools, with built-in Node.js runtime and DeepLink installation.
- MCP Transport Protocol Supports - Connects external tools and data sources via StreamableHTTP, SSE, or Stdio transports using the Model Context Protocol.
- Rich Content Renderers - Displays Markdown, code blocks, images, Mermaid diagrams, and Artifacts, and supports retrying messages and forking conversations.
- Agent Runtime Managers - A tool that selects and manages ACP-compatible coding and task agents as selectable entries within the model selector.
- ACP-Compatible Agent Selectors - Picks ACP-compatible coding and task agents as selectable entries within the model selector for specialized tasks.
- Conversation-Specific Skill Loaders - Loads task instructions, reference files, and optional scripts from folders, ZIP files, or URLs to specialize agent behavior per conversation.
- Skill Packaging - Loading task-specific instructions, references, and scripts from folders, ZIP files, or URLs into conversations for structured workflows.
- Conversation-Specific Skill Loaders - Loads task-specific instructions and scripts from folders, ZIP files, or URLs, and enables them per conversation to specialize agent behavior.
- Graphical Ollama Managers - Downloads, deploys, and runs Ollama models through a graphical interface without requiring command-line operations.
- Deep Link Launchers - Starts conversations or installs MCP services by clicking a link, enabling seamless integration with other applications.
- Web Search Integrations - Integrates search APIs and browser simulation so the model can retrieve and highlight external information during a conversation.
- Agent Session Recorders - Preserving structured work history of agent sessions for recovery, debugging, and future memory flows.
- Chat Content Renderers - Displays Markdown, code blocks, images, Mermaid diagrams, and artifacts within conversations for diverse result presentation.
- Messaging App Bindings - Binds messaging app channels to desktop sessions for creating, switching, and managing conversations remotely.
- Messaging App Session Controllers - Binding messaging app channels to sessions for creating, switching, and managing conversations away from the desktop.
- Agent Skill Packages - Loads task-specific instructions and scripts from external packages into conversations at runtime to specialize agent behavior.
- Agent Session Histories - Preserves structured agent session history with traceable context and tool calls for recovery and debugging.
- AI Chat Clients - Privacy-focused desktop AI assistant with MCP client capabilities.
- Client Applications - Multi-model chat interface with local file and MCP support.
- AI - Listed in the “AI 项目” section of the Great Open Source Project awesome list.
