The AI Coding Agent
That Thinks in Nebulae
Powered by Gemini orchestration. 75+ AI providers. Desktop, CLI, TUI, and Web. A cross-platform agent that reads, writes, and reasons about your code — like a senior engineer that never sleeps.
Everything You Need to Code Faster
A full-featured AI coding agent with intelligent orchestration, multi-model support, and deep integration with your development workflow.
Gemini Orchestration
An intelligent orchestrator that decomposes complex tasks, coordinates multi-step coding workflows, and delegates to specialized sub-agents — all powered by Gemini.
75+ AI Providers
Use any model you want: GPT, Claude, Gemini, DeepSeek, Mistral, local models via Ollama — connect once, switch freely. Provider-agnostic by design.
Cross-Platform
Native desktop app for macOS, Windows, and Linux. Plus a full TUI, CLI, and web interface. One codebase, every surface.
30+ Language Servers
Built-in LSP integration for TypeScript, Python, Go, Rust, Java, C++, Ruby, PHP, and many more. Real code intelligence, not just pattern matching.
Session Management
Multiple concurrent sessions, session sharing via URL, Markdown export, and git-based undo/redo. Your workflow history, always accessible.
MCP Protocol
Extend Nebula X with Model Context Protocol servers. Connect external tools, databases, APIs, and services directly into your coding workflow.
Plugin System
Create custom agents, tools, commands, and skills. Define project-specific behaviors with simple configuration files. Make Nebula X truly yours.
Client/Server Architecture
Run the server anywhere. Attach from any device. Sidecar mode for desktop, headless API server for automation, and remote TUI access.
The Orchestrator
What sets Nebula X apart. An intelligent Gemini-powered orchestration layer that coordinates the entire coding workflow — from understanding your intent to delivering production-ready code.
Your Prompt
Describe what you want to build or fix
Orchestrator
Gemini decomposes the task and plans execution
Sub-Agents
Specialized agents handle each sub-task
Tools
Read, write, bash, LSP, grep, web fetch, and more
Result
Code changes, tests, and explanations delivered
Task Decomposition
Breaks complex requests into manageable sub-tasks
Multi-Agent Delegation
Assigns specialized agents for exploration, planning, and execution
Tool Coordination
Chooses the right tools: file editing, shell commands, LSP, web fetch
Context Management
Maintains deep understanding of your codebase across sessions
See It in Action
Watch Nebula X analyze, plan, and execute — all from your terminal.
Your Models, Your Choice
Provider-agnostic by design. Connect to 75+ AI providers or run local models. Switch models freely without changing your workflow.
OpenAI
GPT-4o, GPT-5, o1
Anthropic
Claude Opus, Sonnet
Gemini Pro, Ultra
Amazon
Bedrock
Azure
OpenAI Service
GitHub
Copilot
DeepSeek
V3, Coder
Mistral
Large, Codestral
Groq
LLaMA, Mixtral
Ollama
Local Models
OpenRouter
Any Model
Perplexity
Sonar
xAI
Grok
Cohere
Command R+
Together
Open Models
75+ More
& Growing
Get Started in Seconds
One command. That's all it takes.
curl -fsSL https://dlmworld.com/install | bash