LLM ReferenceLLM Reference
FrameworkOpen Source

SWE-agent

by Princeton NLP Group
Supports 0 models · Framework · Free open-source

SWE-agent is an open-source agent framework for AI-assisted coding, built by Princeton NLP Group. SWE-agent's verified model list is still being expanded, so this page emphasizes runtime shape, pricing, and comparable AI coding harnesses. It's free and open source. Developers import it as a library to build custom agents, wire model calls to tools, and standardize LLM ops across internal automation.

Overview

Research framework and agent for automated software engineering that interfaces AI models with code repositories using custom terminal-like tools.

Featured Model Specs

No verified underlying model rows yet.

Underlying Models

No verified model links are attached to this harness yet.

Compare to Similar Harnesses

FAQ

What is SWE-agent?

SWE-agent is an open-source agent framework for AI-assisted coding, built by Princeton NLP Group. Developers import it as a library to build custom agents, wire model calls to tools, and standardize LLM ops across internal automation. It is primarily used for LLM ops, custom agent orchestration, and repeatable automation.

How much does SWE-agent cost?

SWE-agent is free and open source. You pay only for the underlying model API usage.

Is SWE-agent open source?

Yes, SWE-agent is open-source software.

What are alternatives to SWE-agent?

Alternatives include Windsurf, Augment Code, and Zed.

Explore Related Pages

Last reviewed: 2026-05-04. Data sourced from harness documentation and provider pages.