---
title: "Pod run record"
method: GET
path: "/pods/{id}/record"
tags: ["Support API Key Authentication"]
---

# Pod run record

`GET /pods/{id}/record`

## Request

- Base URL: `https://lium.io/api`
- URL: `https://lium.io/api/pods/{id}/record`
- Auth: one of:
  - HTTP bearer
  - API key in header `X-API-Key`

One record of a run, to repeat or cite it: GPU model and count, the image reference the container was started from (with its digest when the template pins one), the template, the node's NVIDIA driver, CUDA version and GPU interconnect as the listing showed them when the container came up (`launch_snapshot`), start and end, why it ended, and the total charged (the statement's total). Answers for a removed pod. `launch_snapshot` is null for a run that started before it was recorded. 404 under the same rule as GET /pods/{id}/statement.

## Path parameters

- `id` string, uuid, required

## Response `200`

Run record assembled

- PodRunRecord: What one run ran on, what it cost and when, readable after the pod is removed.
  - `pod_id` string, uuid, required
  - `pod_name` string, nullable, required
  - `removed` boolean, required
  - `gpu_name` string, nullable, required
  - `gpu_count` integer, nullable, required
  - `executor_id` string, uuid, nullable, required: The node (executor.id) the pod ran on last.
  - `image` string, nullable, required: `launch_snapshot.image`; for a run older than the launch snapshot, the image on the rental.
  - `created_at` string, date-time, nullable, required
  - `removed_at` string, date-time, nullable, required
  - `close_reason` string, nullable, required
  - `launch_snapshot` PodLaunchSnapshot, required
    - `template_id` string, uuid, nullable, required: Template the container was started from; null for a Dockerfile.
    - `image` string, nullable, required: Image reference as dispatched: `repo:tag`, or `repo@sha256:…` when the template pins a digest.
    - `gpu_driver` string, nullable, required: NVIDIA driver build on the node, e.g. 570.133.20.
    - `cuda_driver` string, nullable, required: Highest CUDA version that driver supports, major.minor.
    - `gpu_interconnect` string, nullable, required: nvlink when every GPU pair on the node is joined by NVLink, pcie when one is not; null on a single-GPU node or before the topology was scraped.
    - `recorded_at` string, date-time, required: When the container came up and these were recorded.
  - `billed_seconds` number, required
  - `total` number, required: USD charged, the same total GET /pods/{id}/statement returns.
  - `price_segments` PodStatementPriceSegment[], required
    - `price_per_hour` number, required
    - `billed_seconds` number, required
    - `amount` number, required

## Other responses

- `404`: Pod not found
- `422`: Validation Error

## Changes

- **2026-10-19** `0fa4dbe7f52a`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/lium/apis/lium-backend-api/changes/pods/:id/record/get.md)

---

[API](https://skmtc.dev/lium/apis/lium-backend-api.md) · [All operations](https://skmtc.dev/lium/apis/lium-backend-api/llms.txt) · [OpenAPI document](https://skmtc.dev/lium/apis/lium-backend-api/revisions/a682d3ef1e07?raw)