Get a GPU type

GET

catalog

gpus

{id}

Get a GPU type

curl --request GET \
  --url https://api.runpod.io/v2/catalog/gpus/{id} \
  --header 'Authorization: Bearer <token>'

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.runpod.io/v2/catalog/gpus/{id}', 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.runpod.io/v2/catalog/gpus/{id}",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($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.runpod.io/v2/catalog/gpus/{id}"

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

	req.Header.Add("Authorization", "Bearer <token>")

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

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

	fmt.Println(string(body))

}

require 'uri'
require 'net/http'

url = URI("https://api.runpod.io/v2/catalog/gpus/{id}")

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

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body

{
  "id": "NVIDIA GeForce RTX 4090",
  "name": "RTX 4090",
  "pool": "ADA_24",
  "manufacturer": "NVIDIA",
  "memory": 24,
  "secure": true,
  "community": true,
  "price": {
    "secure": 0.44,
    "community": 0.31,
    "serverless": 1.1
  },
  "maxCount": {
    "secure": 8,
    "community": 4
  },
  "availability": "HIGH",
  "dataCenters": [
    {
      "id": "US-KS-2",
      "name": "US Kansas 2",
      "availability": "HIGH"
    }
  ]
}

{
  "title": "Unauthorized",
  "status": 401,
  "detail": "missing bearer token"
}

{
  "title": "Forbidden",
  "status": 403,
  "detail": "access denied"
}

{
  "title": "Not Found",
  "status": 404,
  "detail": "resource not found"
}

{
  "title": "Too Many Requests",
  "status": 429,
  "detail": "rate limit exceeded for the minute window"
}

{
  "title": "Not Found",
  "status": 404,
  "detail": "pod not found",
  "errors": [
    "$: additional properties 'bogus' not allowed"
  ]
}

Authorizations

Authorization

string

header

required

Runpod API key authentication. Generate an API key in the Runpod console and send it in the Authorization header as Bearer <api_key>. Keys are scoped to the permissions granted when created; requests may return 403 when a valid key lacks access to the requested resource or action.

Console sessions (Clerk JWTs) act for the signed-in user. To act for a team instead, also send X-Team-Id: <team_id>; it applies only when the user owns or belongs to that team, otherwise the request runs against the user's own account. API keys created inside a team already act for that team and do not need the header.

Path Parameters

string

required

Query Parameters

include

enum<string>[]

Comma-separated optional expansions. Supported value today: AVAILABILITY. This may expand with more include values in the future.

Maximum array length: 1

Catalog include expansion. Only AVAILABILITY is supported today; additional include values may be added in the future.

Available options:

AVAILABILITY

product

enum<string>[]

Comma-separated availability product contexts. Supported values: POD, CLUSTER, SERVERLESS. Required with include=AVAILABILITY, and valid only with it (400 either way). There is no default: the same GPU type can be scarce for pods and plentiful for serverless, so the context has to be stated rather than assumed.

Catalog product availability context. Availability is product-specific, so this is required whenever availability is requested.

Available options:

POD,

CLUSTER,

SERVERLESS

count

integer

default:1

GPU count for availability and lowest-price calculations. Valid only with include=AVAILABILITY. Defaults to 1.

Required range: x >= 1

cloud

enum<string>

Cloud type for availability and lowest-price calculations. Valid only with include=AVAILABILITY. Supported values: SECURE, COMMUNITY. Default when omitted: SECURE. GPU availability cloud filter.

Available options:

SECURE,

COMMUNITY

countryCodes

string[]

Comma-separated ISO 3166-1 alpha-2 country codes, uppercase, to constrain availability to — e.g. FR or FR,DE. Values within this filter use OR semantics. Valid only with include=AVAILABILITY (400 otherwise); a malformed entry is a 422. Scopes availability, lowest-price calculations and the dataCenters array to those countries, so a listed data center outside them is omitted rather than returned with availability NONE. On the list endpoint a GPU type with no data center in those countries drops out entirely; the single-GPU endpoint still returns the requested type, with availability NONE and dataCenters omitted, so a 404 keeps meaning the GPU type does not exist. Read the NONE on availability rather than the absence of dataCenters, which is also absent when availability was not requested.

Pattern: ^[A-Z]{2}$

cudaVersions

string[]

Comma-separated CUDA versions to scope availability and lowest-price calculations to, matched exactly. Format: major.minor, e.g. 12.8 — a bare major is rejected here because it identifies no version. Valid only with include=AVAILABILITY (400 otherwise) and mutually exclusive with minCudaVersion (400 if both are sent); a malformed entry is a 422. Also narrows the returned cudaVersions array; omit it to enumerate every version offered.

Pattern: ^\d+\.\d+$

minCudaVersion

string

Lowest acceptable CUDA version to scope availability and lowest-price calculations to, compared numerically. Format: integer major or major.minor, e.g. 12 or 12.1 — unlike the gpu.minCudaVersion body field on pod and endpoint create, a bare major is accepted here and means any release of that major, because this filter only widens a read. Valid only with include=AVAILABILITY (400 otherwise) and mutually exclusive with cudaVersions (400 if both are sent); a malformed value is a 422. Use this for an open-ended floor and cudaVersions for an exact set.

Pattern: ^\d+(\.\d+)?$

Response

string

required

Individual GPU type identifier (use for pod creation)

Example:

"NVIDIA GeForce RTX 4090"

name

string

required

Example:

"RTX 4090"

pool

string | null

required

Serverless GPU pool ID (use for serverless endpoint creation). Null if GPU is not in a serverless pool.

Example:

"ADA_24"

manufacturer

enum<string>

required

Canonical GPU hardware manufacturer.

Available options:

NVIDIA,

AMD,

UNKNOWN

memory

integer

required

VRAM in GB

Example:

24

secure

boolean

required

Available on secure cloud

Example:

true

community

boolean

required

Available on community cloud

Example:

true

price

object

required

List price in USD per hour for a single GPU of this type. Pod rates are quoted separately per cloud (secure, community); serverless is the rate for this GPU's pool. In every case the rate for a unit is the figure times gpu.count; the rate actually billed for a pod is reported as cost on the pod itself.

Show child attributes

maxCount

object

required

The largest number of GPUs you can request on a single pod of this type, quoted separately per cloud. A pod runs on one machine, so this is the GPU count of the largest machine of this type Runpod operates in that cloud.

This is a ceiling, not a stock level — it does not mean that many GPUs are free right now. For current availability, request include=AVAILABILITY&product=POD and read availability (overall) or dataCenters (per data center).

Show child attributes

minVcpuCount

integer

Minimum total vCPU count reported across matching machines for the requested count and availability filters. Present only with include=AVAILABILITY and omitted when availability is NONE.

Example:

8

minRam

integer

Minimum total system RAM in GB reported across matching machines for the requested count and availability filters. This is host RAM, not the GPU VRAM reported in memory. Present only with include=AVAILABILITY and omitted when availability is NONE.

Example:

32

availability

enum<string>

Overall GPU availability for the requested product contexts. Present only when requested with include=AVAILABILITY, which also requires product.

Available options:

NONE,

LOW,

MEDIUM,

HIGH

dataCenters

object[]

Per-datacenter GPU availability for the requested product contexts, listing only the datacenters that offer this GPU in the requested configuration. Present only when requested with include=AVAILABILITY, which also requires product, and omitted entirely when the configuration is unavailable everywhere.

Show child attributes

cudaVersions

object[]

CUDA versions offered by machines with this GPU type, each tagged with current capacity. Present only when requested with include=AVAILABILITY, and scoped by the same filters as availability (count, cloud, product, and whichever of cudaVersions / minCudaVersion was supplied).

Machines that report no CUDA version are skipped, so this property is absent entirely for a GPU type with none — AMD, for instance. Treat a missing cudaVersions the same as an empty one. A version absent from a populated list is not offered for this GPU type.

Show child attributes

Last modified on October 25, 2026

Get a GPU type

curl --request GET \
  --url https://api.runpod.io/v2/catalog/gpus/{id} \
  --header 'Authorization: Bearer <token>'

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.runpod.io/v2/catalog/gpus/{id}', 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.runpod.io/v2/catalog/gpus/{id}",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($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.runpod.io/v2/catalog/gpus/{id}"

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

	req.Header.Add("Authorization", "Bearer <token>")

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

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

	fmt.Println(string(body))

}

require 'uri'
require 'net/http'

url = URI("https://api.runpod.io/v2/catalog/gpus/{id}")

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

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body

{
  "id": "NVIDIA GeForce RTX 4090",
  "name": "RTX 4090",
  "pool": "ADA_24",
  "manufacturer": "NVIDIA",
  "memory": 24,
  "secure": true,
  "community": true,
  "price": {
    "secure": 0.44,
    "community": 0.31,
    "serverless": 1.1
  },
  "maxCount": {
    "secure": 8,
    "community": 4
  },
  "availability": "HIGH",
  "dataCenters": [
    {
      "id": "US-KS-2",
      "name": "US Kansas 2",
      "availability": "HIGH"
    }
  ]
}

{
  "title": "Unauthorized",
  "status": 401,
  "detail": "missing bearer token"
}

{
  "title": "Forbidden",
  "status": 403,
  "detail": "access denied"
}

{
  "title": "Not Found",
  "status": 404,
  "detail": "resource not found"
}

{
  "title": "Too Many Requests",
  "status": 429,
  "detail": "rate limit exceeded for the minute window"
}

{
  "title": "Not Found",
  "status": 404,
  "detail": "pod not found",
  "errors": [
    "$: additional properties 'bogus' not allowed"
  ]
}

API v2

Organization

Account

Pods

Clusters

Serverless

Templates

Network Volumes

Global Volumes

Registries

Catalog

Cost Centers

Audit Logs

Billing

Authorizations

Path Parameters

Query Parameters

Response