---
title: "Re-run the final step of a run that stopped on it"
method: POST
path: "/v1/runs/{id}/resume-finish"
tags: ["runs"]
---

# Re-run the final step of a run that stopped on it

`POST /v1/runs/{id}/resume-finish`

## Request

- Base URL: `https://api.hyperresearch.ai`
- URL: `https://api.hyperresearch.ai/v1/runs/{id}/resume-finish`
- Auth: one of:
  - HTTP bearer
  - API key in cookie `hr_session`

For a run whose every step completed and which then failed on `finish`: re-runs `finish` on the SAME run id, against the artifacts already stored. No new run, no new charge. A second call while the resume is in flight answers 202 with `resumed: false` and the live `instance_id`. A resume that itself FAILED can be re-entered, up to three attempts in total (`attempt` says which this is); the fourth answers 409 `run_finish_resume_exhausted`. Any other run state is a 409 `run_not_finish_resumable`.

## Path parameters

- `id` string, required

## Response `202`

Finish re-entered, or already re-entered by an earlier call

- ResumeFinishResponse
  - `id` string, required
  - `status` 'clarifying' | 'queued' | 'running' | 'paused' | 'blocked' | 'done' | 'failed' | 'aborted', required
  - `resumed` boolean, required
  - `attempt` integer
  - `instance_id` string

## Other responses

- `400`: Invalid request
- `401`: Authentication failed
- `403`: Insufficient scope or plan
- `404`: Not found
- `429`: Rate limited or over quota
- `500`: Internal error

---

[API](https://skmtc.dev/hyperresearch/apis/hyperresearch-cloud-api.md) · [All operations](https://skmtc.dev/hyperresearch/apis/hyperresearch-cloud-api/llms.txt) · [OpenAPI document](https://skmtc.dev/hyperresearch/apis/hyperresearch-cloud-api/revisions/726ef831ebf4?raw)