# How to resume a long refinement run after it stops midway with harness2

Repo: [google-research/harness2](https://github.com/google-research/harness2) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Re-issue the same command to continue a multi-task run without redoing finished tasks.

**Who it is for:** Researcher running a larger JobBench or LAB sweep that was interrupted by a crash or quota limit

**The problem:** Long model-backed runs cost money, and restarting from scratch after an interruption wastes both time and API spend.

**The outcome:** A completed run that picked up where it stopped, with results kept under the same runs directory.

## Steps

1. Note the exact command you used, including --bench, --k, --mode, any subset flags and seed.
2. Re-run the identical harness2 command to resume.
3. Use a new HARNESS2_RUNS_DIR only if you changed models or task selection.
4. Run harness2-judge and harness2-report with the same flags once refinement finishes.
5. Check each command's --help for extra options.

## Prompt to paste into your agent

```text
My google-research/harness2 run on <benchmark> stopped partway. Reconstruct the exact original command, re-run it to resume, and verify that previously finished tasks were not repeated by checking the runs/ folder.
```

## Caveat

Resume only works when you repeat the same benchmark, executor, subset, --k, --mode and seed, and changing models or task selection requires a new HARNESS2_RUNS_DIR.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/google-research-harness2/resume-interrupted-multi-task-refinement-run/