AI-powered coding assistant that works with any LLM provider. Command-line interface for querying and editing codebases, generating applications, and automating development workflows.
Get started immediately with powerful LLM options:
# Free Gemini models
/auth gemini enable
/provider gemini
/model gemini-3-flash-preview
# Free Qwen models
/auth qwen enable
/provider qwen
/model qwen-3-coder
# Your Claude Pro / Max subscription
/auth anthropic enable
/provider anthropic
/model claude-sonnet-4-5-20250929
# Your ChatGPT Plus / Pro subscription (Codex)
/auth codex enable
/provider codex
/model gpt-5.2
# Kimi subscription (K2 Thinking with reasoning)
/provider kimi
/key **************
/model kimi-k2-thinking- Use Your Existing Subscriptions: Use Claude Pro/Max, ChatGPT Plus/Pro (Codex) directly via OAuth. Use Kimi/Synthetic/Chutes subscriptions via keys.
- Multi-Account Failover: Configure multiple OAuth accounts that automatically failover on rate limits
- Load Balancer Profiles: Balance requests across providers or accounts with automatic failover
- Free Tier Support: Start coding immediately with Gemini or Qwen free tiers
- Provider Flexibility: Switch between any Anthropic, Gemini, OpenAI, Kimi, or OpenAI-compatible provider
- Top Open Models: Works seamlessly with GLM-4.7, Kimi K2 Thinking, MiniMax M2.1, and Qwen 3 Coder
- Local Models: Run models locally with LM Studio, llama.cpp for complete privacy
- Privacy First: No telemetry by default, local processing available
- Subagent Flexibility: Create agents with different models, providers, or settings
- Interactive REPL: Beautiful terminal UI with multiple themes
- Zed Integration: Native Zed editor integration for seamless workflow
# Install and get started
npm install -g @vybestack/llxprt-code
llxprt
# Try without installing
npx @vybestack/llxprt-code --provider synthetic --model hf:zai-org/GLM-4.7 --keyfile ~/.synthetic_key "simplify the README.md"LLxprt Code is a command-line AI assistant designed for developers who want powerful LLM capabilities without leaving their terminal. Unlike GitHub Copilot or ChatGPT, LLxprt Code works with any provider and can run locally for complete privacy.
Key differences:
- Open source & community driven: Not locked into proprietary ecosystems
- Provider agnostic: Not locked into one AI service
- Local-first: Run entirely offline if needed
- Developer-centric: Built specifically for coding workflows
- Terminal native: Designed for CLI workflows, not web interfaces
- Prerequisites: Node.js 20+ installed
- Install:
npm install -g @vybestack/llxprt-code # Or try without installing: npx @vybestack/llxprt-code - Run:
llxprt - Choose provider: Use
/providerto select your preferred LLM service - Start coding: Ask questions, generate code, or analyze projects
First session example:
cd your-project/
llxprt
> Explain the architecture of this codebase and suggest improvements
> Create a test file for the user authentication module
> Help me debug this error: [paste error message]- Subscription OAuth - Use Claude Pro/Max, ChatGPT Plus/Pro (Codex), or Kimi subscriptions directly
- Free Tiers - Gemini, Qwen free tiers with generous limits
- Multi-Account Failover - Configure multiple OAuth buckets that failover automatically on rate limits
- Load Balancer Profiles - Balance across providers/accounts with roundrobin or failover policies
- Extensive Provider Support - Anthropic, Gemini, OpenAI, Kimi, and any OpenAI-compatible provider Provider Guide →
- Top Open Models - GLM-4.7, Kimi K2 Thinking, MiniMax M2.1, Qwen 3 Coder
- Local Model Support - LM Studio, llama.cpp, Ollama for complete privacy
- Profile System - Save provider configurations and model settings
- Advanced Subagents - Isolated AI assistants with different models/providers
- MCP Integration - Connect to external tools and services
- Beautiful Terminal UI - Multiple themes with syntax highlighting
Interactive Mode (REPL): Perfect for exploration, rapid prototyping, and iterative development:
# Start interactive session
llxprt
> Explore this codebase and suggest improvements
> Create a REST API endpoint with tests
> Debug this authentication issue
> Optimize this database queryNon-Interactive Mode: Ideal for automation, CI/CD, and scripted workflows:
# Single command with immediate response
llxprt --profile-load zai-glm46 "Refactor this function for better readability"
llxprt "Generate unit tests for payment module" > tests/payment.test.jsLLxprt Code works seamlessly with the best open-weight models:
- Context Window: 262,144 tokens
- Architecture: Trillion-parameter MoE (32B active)
- Strengths: Deep reasoning, multi-step tool orchestration, 200-300 sequential tool calls
- Special: Native thinking/reasoning mode with tool interleaving
/provider kimi
/model kimi-k2-thinking
# Or via Synthetic/Chutes:
/provider synthetic
/model hf:moonshotai/Kimi-K2-Thinking- Context Window: 200,000 tokens
- Max Output: 131,072 tokens
- Architecture: Mixture-of-Experts with 355B total parameters (32B active)
- Strengths: Coding, multi-step planning, tool integration
- Context Window: 196,608 tokens
- Architecture: MoE with 230B total parameters (10B active)
- Strengths: Coding workflows, multi-step agents, tool calling
- Cost: Only 8% of Claude Sonnet, ~2x faster
- Context Window: 262,144 tokens
- Max Output: 65,536 tokens
- Architecture: MoE with 480B total parameters (35B active)
- Strengths: Agentic coding, browser automation, tool usage
- Performance: State-of-the-art on SWE-bench Verified (69.6%)
Run models completely offline for maximum privacy:
# With LM Studio
/provider openai
/baseurl http://localhost:1234/v1/
/model your-local-model
# With Ollama
/provider ollama
/model codellama:13bSupported local providers:
- LM Studio: Easy Windows/Mac/Linux setup
- llama.cpp: Maximum performance and control
- Ollama: Simple model management
- Any OpenAI-compatible API: Full flexibility
Create specialized AI assistants with isolated contexts and different configurations:
# Subagents run with custom profiles and tool access
# Access via the commands interface
/subagent list
/subagent create <name>Each subagent can be configured with:
- Different providers (Gemini vs Anthropic vs Qwen vs Local)
- Different models (Flash vs Sonnet vs GLM-4.7 vs Custom)
- Different tool access (Restrict or allow specific tools)
- Different settings (Temperature, timeouts, max turns)
- Isolated runtime context (No memory or state crossover)
Subagents are designed for:
- Specialized tasks (Code review, debugging, documentation)
- Different expertise areas (Frontend vs Backend vs DevOps)
- Tool-limited environments (Read-only analysis vs Full development)
- Experimental configurations (Testing new models or settings)
LLxprt Code integrates with the Zed editor using the Agent Communication Protocol (ACP):
{
"agent_servers": {
"llxprt": {
"command": "/opt/homebrew/bin/llxprt",
"args": ["--experimental-acp", "--profile-load", "my-profile", "--yolo"]
}
}
}Configure in Zed's settings.json under agent_servers. Use which llxprt to find your binary path.
Features:
- In-editor chat: Direct AI interaction without leaving Zed
- Code selection: Ask about specific code selections
- Project awareness: Full context of your open workspace
- Multiple providers: Configure different agents for Claude, OpenAI, Gemini, etc.
** Complete Provider Guide →**
- Settings & Profiles: Fine-tune model parameters and save configurations
- Subagents: Create specialized assistants for different tasks
- MCP Servers: Connect external tools and data sources
- Checkpointing: Save and resume complex conversations
- IDE Integration: Connect to VS Code and other editors
** Full Documentation →**
- From Gemini CLI: Migration Guide
- Local Models Setup: Local Models Guide
- Command Reference: CLI Commands
- Troubleshooting: Common Issues
LLxprt Code does not collect telemetry by default. Your data stays with you unless you choose to send it to external AI providers.
When using external services, their respective terms of service apply:
![@gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=64&v=4){kind=small}
