Skip to main content
GET
/
api
/
v2
/
external
/
inference
/
replica
/
{replica_id}
/
machine-metrics
Get replica machine metrics
curl --request GET \
  --url https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics
import requests

url = "https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics"

response = requests.get(url)

print(response.text)
const options = {method: 'GET'};

fetch('https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));
<?php

$curl = curl_init();

curl_setopt_array($curl, [
CURLOPT_URL => "https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "GET",
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}
package main

import (
"fmt"
"net/http"
"io"
)

func main() {

url := "https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics"

req, _ := http.NewRequest("GET", url, nil)

res, _ := http.DefaultClient.Do(req)

defer res.Body.Close()
body, _ := io.ReadAll(res.Body)

fmt.Println(string(body))

}
HttpResponse<String> response = Unirest.get("https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics")
.asString();
require 'uri'
require 'net/http'

url = URI("https://api.example.com/api/v2/external/inference/replica/{replica_id}/machine-metrics")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)

response = http.request(request)
puts response.read_body
{
  "replicaId": "<string>",
  "hostname": "<string>",
  "timeRange": {
    "start": "<string>",
    "end": "<string>"
  },
  "gpuUtilizationPercent": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuMemoryUtilizationPercent": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuTemperatureCelsius": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuPowerWatt": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuPowerLimitWatt": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuClockSmMhz": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuClockMemMhz": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuPcieThroughputRxBytesPerSec": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "gpuPcieThroughputTxBytesPerSec": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "systemRamTotalBytes": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "systemRamUsedBytes": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ],
  "systemCpuUsagePercent": [
    {
      "timestamp": "<string>",
      "value": 123
    }
  ]
}
{
"detail": [
{
"loc": [
"<string>"
],
"msg": "<string>",
"type": "<string>",
"input": "<unknown>",
"ctx": {}
}
]
}

Path Parameters

replica_id
string
required

Query Parameters

start
string | null

ISO 8601 start timestamp (defaults to replica.created_at)

end
string | null

ISO 8601 end timestamp (defaults to replica.last_health_check if terminated, otherwise now)

step
string
default:15s

Query resolution (e.g., '5s', '15s', '1m')

Response

Successful Response

Resource-agnostic machine-level GPU+system metrics response.

replicaId
string
required
hostname
string
required
timeRange
TimeRange · object
required

Time range for the metrics query.

gpuUtilizationPercent
MetricDataPoint · object[]
required
gpuMemoryUtilizationPercent
MetricDataPoint · object[]
required
gpuTemperatureCelsius
MetricDataPoint · object[]
required
gpuPowerWatt
MetricDataPoint · object[]
required
gpuPowerLimitWatt
MetricDataPoint · object[]
required
gpuClockSmMhz
MetricDataPoint · object[]
required
gpuClockMemMhz
MetricDataPoint · object[]
required
gpuPcieThroughputRxBytesPerSec
MetricDataPoint · object[]
required
gpuPcieThroughputTxBytesPerSec
MetricDataPoint · object[]
required
systemRamTotalBytes
MetricDataPoint · object[]
required
systemRamUsedBytes
MetricDataPoint · object[]
required
systemCpuUsagePercent
MetricDataPoint · object[]
required