Neura Market
GPT Engineer logo

GPT Engineer

Free

Specify what you want it to build, the AI asks for clarification, and then builds it.

FreeFree tier
Type
Open Source
Visit Website GitHub (40 stars)

About GPT Engineer

GPT Engineer is an open-source CLI platform for experimenting with code generation. It allows users to specify software requirements in natural language, then watch as an AI writes and executes the code. The tool also supports improving existing code by providing instructions in a prompt file. Built as a precursor to lovable.dev, GPT Engineer is designed for developers who want to rapidly prototype or generate code through conversational AI. It supports Python 3.10-3.12 and offers features like custom model support (local, Azure, etc.), Docker integration, and a benchmarking utility for testing custom agent implementations against datasets like APPS and MBPP. Note: The repository is archived and read-only.

Key Features

Natural language specification of software
AI writes and executes generated code
Improve existing code by specifying desired changes in a prompt file
CLI platform for experimentation and iteration
Supports custom models (OpenAI, local, Azure, etc.)
Docker support for containerized usage
Benchmarking utility for evaluating custom agent implementations against APPS and MBPP datasets
Supports Python 3.10 - 3.12

Pros & Cons

Pros
  • Free and open-source (MIT license)
  • Low barrier to entry – just write a prompt and run the CLI
  • Ability to improve existing code with iterative prompts
  • Supports multiple Python versions and custom LLM providers
  • Includes benchmarking tools for research and experimentation
Cons
  • Repository is archived and no longer actively maintained
  • Requires an OpenAI API key (or other provider) to function
  • CLI-only with no graphical interface
  • Generated code may need manual refinement and debugging
  • Limited to Python environment (3.10-3.12)

Best For

Rapid prototyping from natural language descriptionsAutomated code generation for small to medium projectsIterative improvement of existing codebasesExperimentation with AI code generation techniquesBenchmarking and comparing custom agents for code generation tasks

FAQ

Is GPT Engineer still maintained?
The repository was archived by the owner on April 22, 2026, and is now read-only. It is no longer actively developed, but the code remains available for use and experimentation.
What do I need to run GPT Engineer?
You need a Python 3.10-3.12 environment, an API key for OpenAI or a compatible model provider, and optionally Docker. Installation is via pip (stable) or Poetry (development).
Can GPT Engineer improve existing code?
Yes. Place a prompt file in the project folder describing the improvements, then run `gpte <project_dir> -i` to let the AI modify the code.