README

The minimal AI software engineering agent

📣 mini-swe-agent now powers Ramp SWE-Bench
📣 mini-swe-agent beats Claude Code and Codex on DeepSWE
📣 Run mini-swe-agent on our new & extremely challenging benchmark, ProgramBench
📣 New tutorial on building minimal AI agents





 [!WARNING]
 This is mini-swe-agent v2. Read the migration guide. For the previous version, check out the v1 branch.

In 2024, we built SWE-bench & SWE-agent and helped kickstart the coding agent revolution.

We now ask: What if our agent was 100x simpler, and still worked nearly as well?

mini is

- Widely adopted: Used by Meta, NVIDIA, Essential AI, IBM, Nebius, Anyscale, Princeton University, Stanford University, and many more.
- Minimal: Just some 100 lines of python for the agent class (and a bit more for the environment,
model, and run script) — no fancy dependencies!
- Performant: Scores 74% on the SWE-bench verified benchmark; starts much faster than Claude Code
- Deployable: Supports local environments, docker/podman, singularity/apptainer, bublewrap, contree, and more
- Compatible: Supports all models via litellm, openrouter, portkey, and more. Support for /completion and /response endpoints, interleaved thinking etc.
- Built by the Princeton & Stanford team behind SWE-bench, SWE-agent, and more
- Tested: 



More motivation (for research)

SWE-agent jump-started the development of AI agents in 2024. Back then, we placed a lot of emphasis on tools and special interfaces for the agent.
However, one year later, as LMs have become more capable, a lot of this is not needed at all to build a useful agent!
In fact, the mini agent

- Does not have any tools other than bash — it doesn't even need to use the tool-calling interface of the LMs.
  This means that you can run it with literally any model. When running in sandboxed environments you also don't need to take care
  of installing a single package — all it needs is bash.
- Has a completely linear history — every step of the agent just appends to the messages and that's it.
  So there's no difference between the trajectory and the messages that you pass on to the LM.
  Great for debugging & fine-tuning.
- Executes actions with subprocess.run — every action is completely independent (as opposed to keeping a stateful shell session running).
  This makes it trivial to execute the actions in sandboxes (literally just switch out subprocess.run with docker exec) and to
  scale up effortlessly. Seriously, this is a big deal, trust me.

This makes it perfect as a baseline system and for a system that puts the language model (rather than
the agent scaffold) in the middle of our attention.
You can see the result on the SWE-bench (bash only) leaderboard, that evaluates the performance of different LMs with mini.




More motivation (as a tool)

Some agents are overfitted research artifacts. Others are UI-heavy frontend monsters.

The mini agent wants to be a hackable tool, not a black box.

- Simple enough to understand at a glance
- 

Top issues

Latest release

What's Changed

 fix(agents): count billed calls that fail to parse against cost_limit by @EnesYilmazcode in https://github.com/SWE-agent/mini-swe-agent/pull/915

Full Changelog: https://github.com/SWE-agent/mini-swe-agent/compare/v2.4.5...v2.4.6