Groq Code CLI

Groq Code CLI is an open-source command-line coding assistant developed by the build-with-groq team. It integrates directly with Groq's high-speed inference engine to provide interactive code generation, refactoring, and terminal-based task execution. Operating directly in the terminal, the CLI leverages Groq-hosted models (such as Llama architectures and specialized models like Kimi K2 Instruct) to deliver low-latency responses. Developers interact with the agent via an interactive prompt where they can ask codebase questions, execute commands, and load context from local repositories. The primary differentiator of Groq Code CLI is the execution speed provided by Groq's LPU hardware. This significantly minimizes waiting time during multi-turn agentic loops and iterative code generation. It also includes commands such as `/init` to automatically generate and load workspace context into the session. The project is open-source under a permissive license and requires a valid Groq API key, with usage costs tied directly to standard Groq API consumption rates.

Tags: cli, code assistant, coding agent, OpenSource, terminal

Visit Groq Code CLI

Key features

  • High-Speed Inference — Leverages Groq's LPU infrastructure for near-instant response times during coding sessions. - Project Context Initialization — Provides a built-in /init command that scans the project directory and generates auto-loaded context for subsequent prompts. - Multiple Model Support — Configurable backend compatible with Groq-hosted open weights models, including Llama and Kimi variants. - Interactive Terminal Workflow — Operates natively in the shell, enabling context switching and direct interaction without leaving the command line. - Open-Source Codebase — Publicly auditable repository allowing custom prompt modifications and local extensions.

Use cases

Use Case: Rapid Code Explanations and Refactoring Developers working in the terminal can pipe code snippets or ask the CLI to explain legacy modules and propose refactored implementations without leaving the shell.

Use Case: Automated Repository Onboarding Using the /init command, developers can generate structural metadata and context files for unfamiliar codebases, allowing the agent to answer architectural questions accurately.

Use Case: Fast Scripting and Test Generation Engineers can rapidly prototype unit tests and automation scripts by issuing iterative prompts that return code output with minimal latency.

Who it is for

Groq Code CLI is designed for developers who prefer keyboard-centric terminal workflows and require ultra-low-latency code completions via Groq's API. It is less suitable for users looking for full IDE graphical integration or agent harnesses with autonomous self-healing multi-tool execution pipelines.

Install guide

Installation Clone the repository and install dependencies locally (Node.js/npm environment): bash git clone https://github.com/build-with-groq/groq-code-cli.git cd groq-code-cli npm install npm link

First Run Export your Groq API key in your shell configuration or current environment before launching the CLI: bash export GROQ_API_KEY="your_groq_api_key_here" groq-code

Example Session Initialize context for your current repository and ask the model to analyze a file: bash # Inside your project directory groq-code > /init > Explain the routing logic in src/router.ts and suggest a unit test

Part of CLI Coding Agents

Categories: AI Coding Agents, Code Assistants, Developer Tools

Related CLI agents

  • Qwen Code — Qwen Code is an open-source terminal-based AI coding assistant maintained by the Qwen team (Alibaba Cloud / QwenLM). Built on…
  • ob-1 — ob-1 is an open-source terminal-based AI coding agent created by Overbrilliant. Designed for developer workflows in the shell, it orchestrates…
  • coro-code — coro-code is an open-source, terminal-based AI coding assistant developed in Rust by Blushyes. Built around a compiled native architecture, it…
  • VTCode — VTCode is an open-source terminal-based AI coding agent built in Rust by Vinh Nguyen (vinhnx). Engineered as a high-performance CLI…
  • Nanocoder — Nanocoder is an open-source terminal-based AI coding assistant developed by the Nano-Collective community. Designed for software developers who prefer working…
  • CodeWhale — CodeWhale is an open-source terminal-based AI coding agent developed in Rust by Hunter Bown. It provides an autonomous and semi-autonomous…
  • AI Tools
  • Categories
  • Industries
  • CLI Coding Agents
  • MCP Servers
  • MCP Categories