Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (llm-supercli) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

LLM SuperCLI

A powerful multi-provider LLM command line interface

Features • Installation • Quick Start • Providers • Commands

Features

Multi-Provider Support - Groq, OpenRouter, Together AI, HuggingFace, Ollama, Gemini, Qwen
Free Tiers Available - Gemini, Qwen, and Ollama work without API keys
Interactive UI - Rich terminal with autocomplete, syntax highlighting, and live streaming
Streaming Responses - Real-time token streaming with reasoning display
Session Management - Save, load, and manage conversation sessions
Tool Support - Built-in file operations, shell commands, and directory navigation
MCP Integration - Model Context Protocol server connections
Multiple Modes - Code, chat, and other operational modes

Installation

Via npm (Recommended)

npm install -g llm-supercli

Via pip

pip install llm-supercli

From Source

git clone https://github.com/alisaalisaalisaalisas/LLM_SuperCLI.git
cd LLM_SuperCLI
pip install -e .

Quick Start

# Start the CLI
llm

# Or use the full command
llm-supercli

Providers

Provider	Models	Auth	Cost
Qwen	coder-model, vision-model	OAuth	Free (2K req/day)
Gemini	gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite	OAuth	Free
Groq	llama-3.3-70b, llama-3.1-8b, mixtral-8x7b, gemma2-9b	API Key	Free tier
Ollama	llama3.2, mistral, codellama, phi3	Local	Free
OpenRouter	claude-3.5-sonnet, gpt-4o, gemini-pro-1.5, llama-3.1-405b	API Key	Paid
Together	llama-3.1-405b, llama-3.1-70b, mixtral-8x22b	API Key	Paid
HuggingFace	llama-3-70b, mixtral-8x7b, gemma-7b	API Key	Paid

Authentication

Qwen (Free - OAuth)

Uses Qwen Code CLI credentials:

npm install -g @anthropic-ai/qwen-code
qwen  # Follow OAuth flow

Credentials stored in ~/.qwen/oauth_creds.json

Free tier: 2,000 requests/day, 60 requests/minute

Gemini (Free - OAuth)

Uses Gemini CLI credentials:

npm install -g @anthropic-ai/gemini-cli
gemini  # Follow OAuth flow

Credentials stored in ~/.gemini/oauth_creds.json

Ollama (Free - Local)

Run models locally with Ollama:

# Install Ollama from https://ollama.ai
ollama pull llama3.2

API Keys

Set environment variables for API key providers:

export GROQ_API_KEY=your_key
export OPENROUTER_API_KEY=your_key
export TOGETHER_API_KEY=your_key
export HF_API_KEY=your_key

Or set via CLI:

/key groq your_api_key

Commands

Command	Description
`/help`	Show help message
`/model`	Switch LLM model or provider
`/settings`	View/edit settings
`/login [provider]`	Login with OAuth
`/logout [provider]`	Logout from provider
`/sessions`	Manage chat sessions
`/new`	Start new chat session
`/clear`	Clear the screen
`/status`	Show current status
`/cost`	Show token usage
`/key [provider] [key]`	Set API key
`/mcp`	Manage MCP servers
`/account`	View account information
`/favorite`	Save session as favorite
`/compress`	Compress conversation context
`/rewind`	Rewind to previous message
`/quit`	Exit the CLI
`/update`	Check for updates and install

Updating

From within the CLI

/update

This will check for new versions and prompt you to update if available.

Manual update

# npm installation
npm update -g llm-supercli

# pip installation
pip install --upgrade llm-supercli

After updating, restart the CLI to use the new version.

Special Syntax

# Execute shell command
!ls -la
!git status

# Include file contents in prompt
@path/to/file.py
Tell me about @src/main.py

# Multiple files
Explain the difference between @file1.py and @file2.py

Built-in Tools

The LLM can use these tools during conversations:

Tool	Description
`list_directory(path)`	List files and folders
`read_file(path)`	Read file contents
`write_file(path, content)`	Create or write a file
`create_directory(path)`	Create a folder
`run_command(command)`	Run shell command
`get_current_directory()`	Get current working directory

Configuration

Settings stored in ~/.llm_supercli/config.json

Access interactive settings menu:

/settings

Available settings:

LLM: Provider, model, temperature, max_tokens
UI: Theme, streaming, syntax highlighting
MCP: Enable/disable MCP features

Requirements

Python 3.10+
Node.js 14+ (for npm installation)

Dependencies:

rich >= 13.0.0
httpx >= 0.25.0
click >= 8.0.0
prompt_toolkit >= 3.0.0
packaging >= 21.0

License

MIT License - see LICENSE for details.

llm-supercli