# tbs-protocol-20261010.md

Source: https://www.terminal-bench-science.ai/run
Retrieved: 2026-10-10

# Run Terminal-Bench-Science

How to run Terminal-Bench-Science0.1

## [Install Harbor](https://www.terminal-bench-science.ai/run\#install-harbor)

Terminal-Bench-Science runs in [Harbor](https://www.harborframework.com/).

uvpip

```
uv tool install 'harbor[modal]'
```

## [Run Terminal-Bench-Science](https://www.terminal-bench-science.ai/run\#run-terminal-bench-science)

Terminal-Bench-Science 0.1 contains 70 tasks across the life, physical, earth, mathematical, and engineering sciences. We use Modal for our experiments, and you can also run the benchmark with other Harbor-supported environments, including Daytona. This release does not require GPUs.

```
harbor run -d terminal-bench-science/terminal-bench-science@v0.1 \
  -e modal \
  -a claude-code \
  -m anthropic/claude-opus-5 \
  -n 32
```

## [Analyze results](https://www.terminal-bench-science.ai/run\#analyze-results)

You can run the Harbor viewer if you want to inspect your experiment results.

```
harbor view jobs
```

## [Upload results](https://www.terminal-bench-science.ai/run\#upload-results)

You can also upload your results to [Harbor Hub](https://hub.harborframework.com/) to store for later or share with others. All uploads are private by default.

```
harbor upload jobs/<name> # or harbor run ... --upload
```

## [Join our community](https://www.terminal-bench-science.ai/run\#join-our-community)

If you enjoy the benchmark or have feedback on how we can improve it, [join our Discord](https://discord.com/invite/2Pe5uWGcV3) and let us know in `#tb-science`, or contact [Steven Dillmann](mailto:stevendi@stanford.edu).

## [Citation](https://www.terminal-bench-science.ai/run\#citation)

If you find this work useful, please cite it. The DOI [10.5281/zenodo.22110253](https://doi.org/10.5281/zenodo.22110253) is the concept DOI and always resolves to the latest release; to cite a specific release, use that version’s DOI from the [Zenodo record](https://zenodo.org/records/22110253).

APA

```
Terminal-Bench-Science Team. (2026). Terminal-Bench-Science: Evaluating AI agents on research workflows across scientific domains [Computer software]. https://doi.org/10.5281/zenodo.22110253
```

Copy

BibTeX

```
@software{Terminal-Bench-Science_2026,
  author = {{Terminal-Bench-Science Team}},
  doi = {10.5281/zenodo.22110253},
  license = {Apache-2.0},
  title = {{Terminal-Bench-Science: Evaluating AI agents on research workflows across scientific domains}},
  url = {https://github.com/harbor-framework/terminal-bench-science},
  year = {2026}
}
```

Copy

Fixed release:
[Skip to content](https://github.com/harbor-framework/terminal-bench-science/releases/tag/v0.1.0#start-of-content)

You signed in with another tab or window. [Reload](https://github.com/harbor-framework/terminal-bench-science/releases/tag/v0.1.0) to refresh your session.You signed out in another tab or window. [Reload](https://github.com/harbor-framework/terminal-bench-science/releases/tag/v0.1.0) to refresh your session.You switched accounts on another tab or window. [Reload](https://github.com/harbor-framework/terminal-bench-science/releases/tag/v0.1.0) to refresh your session.Dismiss alert

{{ message }}

# v0.1.0

[Latest](https://github.com/harbor-framework/terminal-bench-science/releases/latest)

[Latest](https://github.com/harbor-framework/terminal-bench-science/releases/latest)

Compare

# Choose a tag to compare

## Sorry, something went wrong.

Filter

Loading

## Sorry, something went wrong.

### Uh oh!

There was an error while loading. [Please reload this page](https://github.com/harbor-framework/terminal-bench-science/releases/tag/v0.1.0).

## No results found

[View all tags](https://github.com/harbor-framework/terminal-bench-science/tags)

![@StevenDillmann](https://avatars.githubusercontent.com/u/168044879?s=40&v=4)[StevenDillmann](https://github.com/StevenDillmann)

released this

Aug 26
26 Aug 10:24


[v0.1.0](https://github.com/harbor-framework/terminal-bench-science/tree/v0.1.0)

[`f81afac`](https://github.com/harbor-framework/terminal-bench-science/commit/f81afac4f11048e77a15dfc8fb1dbfb897fea0ce)

This commit was created on GitHub.com and signed with GitHub’s **verified signature**.


GPG key ID: B5690EEEBB952194

Verified
on Aug 26, 2026, 06:23 AM

[Learn about vigilant mode](https://docs.github.com/github/authenticating-to-github/displaying-verification-statuses-for-all-of-your-commits).


# Terminal-Bench-Science v0.1.0

Initial release containing 70 expert-curated research workflow tasks across the

life, physical, earth, mathematical, and engineering sciences.

DOI: [10.5281/zenodo.22110254](https://doi.org/10.5281/zenodo.22110254)