---
title: "List Gpus"
method: GET
path: "/api/v1/gpus"
tags: ["gpus"]
---

# List Gpus

`GET /api/v1/gpus`

List all GPUs in the cluster with their current metrics.

Returns real-time GPU metrics from DCGM Exporter via Prometheus:
- GPU utilization percentage
- Memory usage (used, free, total in bytes)
- Temperature in Celsius
- Power consumption in Watts

## Response `200`

Successful Response

- GPUSummary
  - `total_gpus` integer, required
  - `gpus` GPUMetrics[], required
    - `gpu_index` integer, required
    - `node` string, required
    - `gpu_uuid` string, nullable
    - `gpu_name` string, nullable
    - `utilization_percent` number, nullable
    - `memory_used_bytes` integer, nullable
    - `memory_free_bytes` integer, nullable
    - `memory_total_bytes` integer, nullable
    - `peak_memory_used_bytes` integer, nullable
    - `temperature_celsius` number, nullable
    - `power_usage_watts` number, nullable

---

[API](https://skmtc.net/nebius/apis/nim-metadata-service.md) · [All operations](https://skmtc.net/nebius/apis/nim-metadata-service/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/nebius/nim-metadata-service/versions/6ed06460dc5e/schema)