# mini-swe-agent — AI Coding Tool Review & Pricing

> mini-swe-agent is a minimal AI software engineering agent from the team behind SWE-bench and SWE-agent. The agent class is about 100 lines of Python, it uses only bash instead of a custom tool interface, and it keeps a linear message history that is also the trajectory, so it is easy to read, modify and run with any model.

Source: https://ai.dosa.dev/tools/mini-swe-agent · Updated Nov 8, 2026 · Category: Autonomous & Async Agents · Published by dosa.dev

## At a glance

- **Tool**: mini-swe-agent
- **Company**: SWE-agent (Princeton & Stanford)
- **Category**: [Autonomous & Async Agents](https://ai.dosa.dev/category/async-agents)
- **Pricing model**: Open Source
- **Official website**: https://mini-swe-agent.com

## Key features

- About 100 lines of Python for the core agent class, built to be hacked on
- Bash as the only tool, so no tool-calling API is required and any model works
- Model support via litellm, OpenRouter and Portkey, including /completion and /response endpoints
- Local, Docker/Podman, Singularity/Apptainer and Bubblewrap execution environments
- `mini` CLI for interactive use plus batch runs over SWE-bench and a trajectory browser
- Install with `uvx mini-swe-agent`, `pipx run mini-swe-agent` or `pip install mini-swe-agent`

## Pricing

**Model**: Open Source

Free and open-source under the MIT License; you pay only for the model API calls (via litellm, OpenRouter, Portkey or a local model) and any sandbox infrastructure you run it in.

## Who it is for

- **Best for**: Researchers and engineers who want a small, transparent coding agent baseline for benchmarks, fine-tuning or sandboxed batch runs, and developers who want a simple bash-only agent in the terminal.
- **Not ideal for**: Developers who want a polished IDE experience, rich built-in tooling or a managed product with team features.

## Verdict

The reference minimal coding agent: tiny, model agnostic and benchmark proven, best as a hackable baseline rather than a full daily driver.

## Recent updates

mini-swe-agent v2 is the current major version (v2.4.6 released Jul 23, 2026); the README now highlights that it powers Ramp SWE-Bench and serves as an evaluation harness for the DeepSWE benchmark, and the original SWE-agent is in maintenance mode with mini-swe-agent as its successor.

## Frequently asked questions

### Is mini-swe-agent free?

Yes - mini-swe-agent is Open Source. Free and open-source under the MIT License; you pay only for the model API calls (via litellm, OpenRouter, Portkey or a local model) and any sandbox infrastructure you run it in.

### Is mini-swe-agent open source?

Yes - mini-swe-agent is open source. Free and open-source under the MIT License; you pay only for the model API calls (via litellm, OpenRouter, Portkey or a local model) and any sandbox infrastructure you run it in.

### Who is mini-swe-agent best for?

Researchers and engineers who want a small, transparent coding agent baseline for benchmarks, fine-tuning or sandboxed batch runs, and developers who want a simple bash-only agent in the terminal.

### Who is mini-swe-agent not ideal for?

Developers who want a polished IDE experience, rich built-in tooling or a managed product with team features.

### What are the best mini-swe-agent alternatives?

The closest mini-swe-agent alternatives on ai.dosa.dev are Jules by Google, OpenHands, SWE-agent - all listed under Autonomous & Async Agents.

### Who makes mini-swe-agent?

mini-swe-agent is developed by SWE-agent (Princeton & Stanford). It is listed in the Autonomous & Async Agents category on ai.dosa.dev.

## Top alternatives

- [Jules by Google](https://ai.dosa.dev/tools/jules) (Freemium) — Developers who want to offload repetitive, non-urgent tasks like dependency version bumps, lint fixes, test scaffolding, and resolving CI failures in GitHub-hosted projects.
- [OpenHands](https://ai.dosa.dev/tools/openhands) (Freemium) — Individual developers and software teams looking to automate repetitive coding tasks, repository maintenance, and issue resolution through autonomous agents.
- [SWE-agent](https://ai.dosa.dev/tools/swe-agent) (Open Source) — Autonomous resolution of software engineering issues in real GitHub repositories and performing custom developer tasks using language models.
- [Sweep](https://ai.dosa.dev/tools/sweep) (Freemium) — Individual developers and teams working primarily in JetBrains IDEs who want a native-feeling assistant for routine coding tasks, bug fixes, and autocompletions.
- [Ellipsis](https://ai.dosa.dev/tools/ellipsis) (Freemium) — Teams looking to centralize and govern AI coding agents across repositories with secure, non-personal credentials and persistent cloud-based execution.
- [Cosine](https://ai.dosa.dev/tools/genie) (Freemium) — Engineering teams performing large-scale code refactors, library migrations, and complex multi-file tasks that require autonomous, background execution.
- Full alternatives guide: https://ai.dosa.dev/tools/mini-swe-agent/alternatives