Console

You can view GPU VM, H4D VM, or TPU future availability in a region when youcreate a future reservation request in calendar mode. For more information,seeCreate a request for GPU VMs, H4D VMs, or TPUs in thisdocument.

gcloud

To view GPU VM, H4D VM, or TPU future availability in a region, use thegcloud beta compute advice calendar-mode command.Based on the resources that you want to view, include the following flags:

• To view GPU VM or H4D VM availability, include the--vm-count and--machine-type flags:

  gcloud beta compute advice calendar-mode \--vm-count=NUMBER_OF_VMS \--machine-type=MACHINE_TYPE \    --region=REGION \    --start-time-range=from=FROM_START_TIME,to=TO_START_TIME \    --end-time-range=from=FROM_END_TIME,to=TO_END_TIME

• To view TPU availability, include the--chip-count and--tpu-versionflags:

  gcloud beta compute advice calendar-mode \--chip-count=NUMBER_OF_CHIPS \--tpu-version=TPU_VERSION \    --region=REGION \    --start-time-range=from=FROM_START_TIME,to=TO_START_TIME \    --end-time-range=from=FROM_END_TIME,to=TO_END_TIME

Replace the following:

• NUMBER_OF_VMS: the number of VMs to reserve. Thevalue must be at least1 and no greater than80 for GPU VMs or256 for H4D VMs.

• MACHINE_TYPE: the GPU or H4D machine type to reserve.Specify one of the following values:

  • For A4 machine types:a4-highgpu-8g

  • For A3 Ultra machine types:a3-ultragpu-8g

  • For A3 Mega machine types:a3-megagpu-8g

  • For A3 High machine types with 8 GPUs:a3-highgpu-8g

  • For H4D HPC machine types: SeeH4D machine types

• NUMBER_OF_CHIPS: the number of TPU chips to reserve.The value must be1,4,8,16,32,64,128,256,512,or1024.

• TPU_VERSION: the TPU version to reserve. Specify oneof the following values:

  • For TPU7x:TPU7X

  • For TPU v6e:V6E

  • For TPU v5p:V5P

  • For TPU v5e:V5E

  If you specify a TPU v5e, then you must include the--workload-type flag.Set this flag to the type of workloads that you want to run on the TPUs:

  • For workloads that handle large amounts of data in single ormultiple operations, such as machine learning (ML) trainingworkloads, specifyBATCH.

  • For workloads that handle concurrent requests and require minimalnetwork latency, such as ML inference workloads, specifySERVING.

    Note: If you specifySERVING, then you can only reserve 1, 4, or 8TPU v5e chips.

• REGION: the region where to reserve GPU VMs, H4D VMs,or TPUs. To check which regions and zones are supported, seeLimitations.

• FROM_START_TIME andTO_START_TIME:the earliest and latest dates that you want to reserve capacity on.Format these dates as RFC 3339 timestamps:

  YYYY-MM-DDTHH:MM:SSOFFSET

  Replace the following:

  • YYYY-MM-DD: a date formatted as a four-digityear, two-digit month, and a two-digit day, separated by hyphens(-).

  • HH:MM:SS: a time formatted as a two-digit hourusing a 24-hour time, two-digit minutes, and two-digit seconds,separated by colons (:).

  • OFFSET: the time zone formatted as an offset ofCoordinated Universal Time (UTC).For example, to use the Pacific Standard Time (PST), specify-08:00. To use no offset, specifyZ.

• FROM_END_TIME andTO_END_TIME: theearliest and latest dates that you want your capacity reservation to endon. Format these dates as RFC 3339 timestamps. If you want to specify arange of durations for your reservation period instead of end times,then replace the--end-time-range flag with the--duration-range flag.

The output is similar to the following:

- recommendationsPerSpec:    spec:endTime: '2026-02-10T00:00:00Z'location: zones/us-central1-a      otherLocations:        zones/us-central1-b:          details: Recommendation in this zone is possible.          status: RECOMMENDED        zones/us-central1-c:          details: Temporarily no free capacity in this zone in the          requested time window.          status: NO_CAPACITY        zones/us-central1-f:          details: This machine family is not supported in this zone.          status: NOT_SUPPORTED      recommendationId: 0d3f005d-f952-4fce-96f2-6af25e1591eb      recommendationType: FUTURE_RESERVATIONstartTime: '2026-02-07T00:00:00Z'

If your requested resources are available, then the output contains thestartTime,endTime, andlocation fields. These fields specify theearliest start time, the latest end time, and the zones when resources areavailable.

REST

To view GPU VM, H4D VM, or TPU future availability in a region, make aGETrequest to thebetaadvice.calendarMode method.Based on the resources that you want to view, include the following fieldsin the request body:

• To view GPU VM or H4D VM availability, include theinstanceCount andmachineType fields:

  POST https://compute.googleapis.com/compute/beta/projects/PROJECT_ID/regions/REGION/advice/calendarMode{  "futureResourcesSpecs": {    "spec": {      "targetResources": {        "specificSkuResources": {"instanceCount": "NUMBER_OF_VMS","machineType": "MACHINE_TYPE"        }      },      "timeRangeSpec": {        "startTimeNotEarlierThan": "FROM_START_TIME",        "startTimeNotLaterThan": "TO_START_TIME",        "endTimeNotEarlierThan": "FROM_END_TIME",        "endTimeNotLaterThan": "TO_END_TIME"      }    }  }}

• To view TPU availability, include theacceleratorCount andvmFamilyfields:

  POST https://compute.googleapis.com/compute/beta/projects/PROJECT_ID/regions/REGION/advice/calendarMode{  "futureResourcesSpecs": {    "spec": {      "targetResources": {        "aggregateResources": {"acceleratorCount": "NUMBER_OF_CHIPS","vmFamily": "TPU_VERSION"        }      },      "timeRangeSpec": {        "startTimeNotEarlierThan": "FROM_START_TIME",        "startTimeNotLaterThan": "TO_START_TIME",        "endTimeNotEarlierThan": "FROM_END_TIME",        "endTimeNotLaterThan": "TO_END_TIME"      }    }  }}

Replace the following:

• PROJECT_ID: the ID of the project where you want toreserve resources.

• REGION: the region where you want to reserve GPU VMs,H4D VMs, or TPUs. To check the regions and zones that are supported, seeLimitations.

• NUMBER_OF_VMS: the number of GPU or H4D VMs toreserve. For GPU VMs, the value must be between1 and80.For H4D VMs, the value must be between1 and256.

• MACHINE_TYPE: the GPU or H4D machine type to reserve.Specify one of the following values:

  • For A4 machine types:a4-highgpu-8g

  • For A3 Ultra machine types:a3-ultragpu-8g

  • For A3 Mega machine types:a3-megagpu-8g

  • For A3 High machine types with 8 GPUs:a3-highgpu-8g

  • For H4D HPC machine types: SeeH4D machine types

• NUMBER_OF_CHIPS: the number of TPU chips to reserve.The value must be1,4,8,16,32,64,128,256,512,or1024.

• TPU_VERSION: the TPU version to reserve. Specify oneof the following values:

  • For TPU7x:VM_FAMILY_CLOUD_TPU_POD_SLICE_TPU7X

  • For TPU v6e:VM_FAMILY_CLOUD_TPU_LITE_POD_SLICE_CT6E

  • For TPU v5p:VM_FAMILY_CLOUD_TPU_POD_SLICE_CT5P

  • For TPU v5e:VM_FAMILY_CLOUD_TPU_LITE_POD_SLICE_CT5LP

  If you specify a TPU v5e, then, in theaggregateResources field, youmust include theworkloadType field.Set this field to the type of workloads that you want to run on theTPUs:

  • For workloads that handle large amounts of data in single ormultiple operations, such as machine learning (ML) trainingworkloads, specifyBATCH.

  • For workloads that handle concurrent requests and require minimalnetwork latency, such as ML inference workloads, specifySERVING.

    Note: If you specifySERVING, then you can only reserve 1, 4, or 8TPU v5e chips.

• FROM_START_TIME andTO_START_TIME:the earliest and latest dates that you want to reserve capacity on.Format these dates as RFC 3339 timestamps:

  YYYY-MM-DDTHH:MM:SSOFFSET

  Replace the following:

  • YYYY-MM-DD: a date formatted as a 4-digityear, 2-digit month, and a 2-digit day, separated by hyphens(-).

  • HH:MM:SS: a time formatted as a 2-digit hourusing a 24-hour time, 2-digit minutes, and 2-digit seconds,separated by colons (:).

  • OFFSET: the time zone formatted as an offset ofCoordinated Universal Time (UTC).For example, to use the Pacific Standard Time (PST), specify-08:00. To use no offset, specifyZ.

• FROM_END_TIME andTO_END_TIME: theearliest and latest dates that you want your capacity reservation to endon. Format these dates as RFC 3339 timestamps. If you want to specify arange of durations for your reservation period instead of end times,then replace theendTimeNotEarlierThan andendTimeNotLaterThanfields with theminDuration andmaxDuration fields.

The output is similar to the following:

{  "recommendations": [    {      "recommendationsPerSpec": {        "spec": {          "recommendationId": "a21a2fa0-72c7-4105-8179-88de5409890b",          "recommendationType": "FUTURE_RESERVATION","startTime": "2026-02-07T00:00:00Z","endTime": "2026-02-10T00:00:00Z",          "otherLocations": {            "zones/us-central1-b": {              "status": "RECOMMENDED",              "details": "Recommendation in this zone is possible."            },            "zones/us-central1-c": {              "status": "NO_CAPACITY",              "details": "Temporarily no free capacity in this zone in the                          requested time window."            },            "zones/us-central1-f": {              "status": "NOT_SUPPORTED",              "details": "This machine family is not supported in this zone."            }          },"location": "zones/us-central1-a"        }      }    }  ]}

If your requested resources are available, then the output contains thestartTime,endTime, andlocation fields. These fields specify theearliest start time, the latest end time, and the zones when resources areavailable.

Console

1. In the Google Cloud console, go to theReservations page.

   Go to Reservations

2. Click theFuture reservations tab.

3. Clickadd_boxCreate future reservation. TheCreate a future reservation pageappears and theHardware configuration pane is selected.

4. In theConfigurations section, specify the properties of theGPU VMs, H4D VMs, or TPUs that you want to reserve by doing one of thefollowing:

   • To specify GPU VM, H4D VM, or TPU properties directly, complete thefollowing steps:

     1. SelectSpecify machine type.

     2. Click theGPUs,TPUs, orCompute-optimized tab, andthen select asupportedGPU machine type, H4D machine type, or TPU version.

   • To specify GPU or H4D VM properties by using an existing VM asreference, selectUse existing VM, and then select the VM.

5. If you specified a TPU v5e (CT5LP) in the previous step, then, intheTPU v5 workload type list, select one of the following options:

   • To run workloads on the TPUs that handle large amounts of data insingle or multiple operations, such as ML training workloads, selectBatch.

   • To run workloads on the TPUs that handle concurrent requests andrequire minimal network latency, such as ML inference workloads,selectServing.

     Note: If you selectServing, then you can only reserve 1, 4, or8 TPU v5e chips.

6. In theSearch for capacity section, complete the following steps:

   1. In theRegion andZone lists, select the region and zonewhere you want to reserve resources.

   2. In theTotal capacity needed field (when reserving GPU or H4DVMs) orNumber of chips list (when reserving TPUs), specify thenumber of GPU VMs, H4D VMs, or TPU chips to reserve. You can specifythe following values:

      • For GPU VMs, a value between1 and80

      • For H4D VMs, a value between1 and256

      • For TPU chips, a value of1,4,8,16,32,64,128,256,512, or1024

   3. In theStart time list, select the start time for your request.

   4. Optional: In theChoose your start date flexibility list, selecthow exact your start date needs to be.

   5. In theReservation duration field, specify for how long you wantto reserve resources.

   6. ClickSearch for capacity. Then, in theAvailable capacitytable, select one of the available options that contain the type,number, and reservation period of the GPU VMs, H4D VMs, or TPUs toreserve.

7. ClickNext.

8. In theShare type section, select the projects to share yourrequested capacity with:

   • To use the reserved capacity only within your project, selectLocal.

   • To share the reserved capacity with other projects, selectShared, clickaddAdd projects, and then follow the prompts to select theprojects.

   Important: You can only specify the share type and shared projects whenyou create a request. You can't modify these settings after submission.

9. ClickNext.

10. In theFuture reservation name field, enter a name for the request.

11. In theReservation name field, enter the name of the reservationthat Compute Engine automatically creates to provision yourrequested capacity.

12. ClickCreate.

gcloud

To create a future reservation request in calendar mode and submit it forreview, use thegcloud compute future-reservations create command.Based on the resources that you want to reserve, include the followingflags:

• To reserve GPU or H4D VMs, include the--total-count,--machine-type, and--deployment-type=DENSE flags:

  gcloud compute future-reservations createFUTURE_RESERVATION_NAME \    --auto-delete-auto-created-reservations \--total-count=NUMBER_OF_VMS \--machine-type=MACHINE_TYPE \--deployment-type=DENSE \    --planning-status=SUBMITTED \    --require-specific-reservation \    --reservation-mode=CALENDAR \    --reservation-name=RESERVATION_NAME \    --share-setting=SHARE_TYPE \    --start-time=START_TIME \    --end-time=END_TIME \    --zone=ZONE

• To reserve TPUs, include the--chip-count and--tpu-version flags:

  gcloud compute future-reservations createFUTURE_RESERVATION_NAME \    --auto-delete-auto-created-reservations \--chip-count=NUMBER_OF_CHIPS \--tpu-version=TPU_VERSION \    --planning-status=SUBMITTED \    --require-specific-reservation \    --reservation-mode=CALENDAR \    --reservation-name=RESERVATION_NAME \    --share-setting=SHARE_TYPE \    --start-time=START_TIME \    --end-time=END_TIME \    --zone=ZONE

Replace the following:

• FUTURE_RESERVATION_NAME: the name of the request.

• NUMBER_OF_VMS: the number of GPU or H4D VMs toreserve:

  • For GPU VMs, the value must be between1 and80.

  • For H4D VMs, the value must be between1 and256.

  Specify a number of VMs that is equal or lower than the number of VMsthat youconfirmed as available.

• MACHINE_TYPE: the GPU or H4D machine type to reserve.

• NUMBER_OF_CHIPS: the number of TPU chips to reserve.Specify a value equal to or lower than the number of chips that youconfirmed as available. The value must be one ofthe following:1,4,8,16,32,64,128,256,512, or1024.

• TPU_VERSION: the TPU version to reserve. If youspecify a TPU v5e (V5E), then you must include the--workload-type flagto specify type of workloads to run on the TPUs.

• RESERVATION_NAME: the name of the reservation thatCompute Engine automatically creates to provision your requestedcapacity.

• SHARE_TYPE: whether other projects in yourorganization can consume the reserved capacity. Specify one of thefollowing values:

  • To use capacity only within your project:local

  • To share capacity with other projects:projects

  If you specifyprojects, then you must include the--share-with flagset to a comma-separated list of project IDs—for example,project-1,project-2. You can specify up to 100 projects within yourorganization. Don't include your project ID in this list. You canconsume the reserved capacity by default.

  Important: You can only specify the share type and shared projects whenyou create a request. You can't modify these settings after submission.

• START_TIME: the start time of the request, which youmust format as an RFC 3339 timestamp.

• END_TIME: the end time of your reservation period,which you must format as an RFC 3339 timestamp. If you want to specify aduration, in seconds, for your reservation period instead of an endtime, then replace the--end-time flag with the--duration flag.

• ZONE: the zone where you want to reserve resources.

REST

To create a future reservation request in calendar mode and submit it forreview, make aPOST request to thefutureReservations.insert method.Based on the resources that you want to reserve, include the followingfields in the request body:

• To reserve GPU or H4D VMs, include thetotalCount andmachineTypefields, as well as thedeploymentType field set toDENSE:

  POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/futureReservations{  "name": "FUTURE_RESERVATION_NAME",  "autoDeleteAutoCreatedReservations": true,"deploymentType": "DENSE",  "planningStatus": "SUBMITTED",  "reservationMode": "CALENDAR",  "reservationName": "RESERVATION_NAME",  "shareSettings": {    "shareType": "SHARE_TYPE"  },  "specificReservationRequired": true,  "specificSkuProperties": {"totalCount":NUMBER_OF_VMS,"instanceProperties": {      "machineType": "MACHINE_TYPE"    }  },  "timeWindow": {    "startTime": "START_TIME",    "endTime": "END_TIME"  }}

• To reserve TPUs, include theacceleratorCount andvmFamily fields:

  POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/futureReservations{  "name": "FUTURE_RESERVATION_NAME",  "autoDeleteAutoCreatedReservations": true,  "planningStatus": "SUBMITTED",  "reservationMode": "CALENDAR",  "reservationName": "RESERVATION_NAME",  "shareSettings": {    "shareType": "SHARE_TYPE"  },  "specificReservationRequired": true,  "aggregateReservation": {    "reservedResources": [      {        "accelerator": {"acceleratorCount":NUMBER_OF_CHIPS        }      }    ],"vmFamily": "TPU_VERSION"  },  "timeWindow": {    "startTime": "START_TIME",    "endTime": "END_TIME"  }}

Replace the following:

• PROJECT_ID: the ID of the project where you want tocreate the request.

• ZONE: the zone where you want to reserve resources.

• FUTURE_RESERVATION_NAME: the name of the request.

• RESERVATION_NAME: the name of the reservation thatCompute Engine automatically creates to provision your requestedcapacity.

• SHARE_TYPE: whether other projects in yourorganization can consume the reserved capacity. Specify one of thefollowing values:

  • To use capacity only within your project:LOCAL

  • To share capacity with other projects:SPECIFIC_PROJECTS

  If you specifySPECIFIC_PROJECTS, then, in theshareSettings field,you must include theprojectMap fieldto specify the projects to share the capacity with. You can specify upto 100 projects within your organization. Don't specify your project ID.You can consume the reserved capacity by default.

  Important: You can only specify the share type and shared projects whenyou create a request. You can't modify these settings after submission.

  For example, to share the requested capacity with two other projects,include the following:

  "shareSettings": {  "shareType": "SPECIFIC_PROJECTS","projectMap": {    "CONSUMER_PROJECT_ID_1": {      "projectId": "CONSUMER_PROJECT_ID_1"    },    "CONSUMER_PROJECT_ID_2": {      "projectId": "CONSUMER_PROJECT_ID_2"    }  }}

  ReplaceCONSUMER_PROJECT_ID_1 andCONSUMER_PROJECT_ID_2 with the IDs of two projectsthat you want to allow to consume the requested capacity.

• NUMBER_OF_VMS: the number of GPU or H4D VMs toreserve:

  • For GPU VMs, the value must be between1 and80.

  • For H4D VMs, the value must be between1 and256.

  Specify a number of VMs that is equal or lower than the number of VMsthat youconfirmed as available.

• MACHINE_TYPE: the GPU or H4D machine type to reserve.

• NUMBER_OF_CHIPS: the number of TPU chips to reserve.Specify a value equal to or lower than the number of chips that youconfirmed as available. The value must be one ofthe following:1,4,8,16,32,64,128,256,512, or1024.

• TPU_VERSION: the TPU version to reserve. If youspecify a TPU v5e (VM_FAMILY_CLOUD_TPU_LITE_POD_SLICE_CT5LP), then, intheaggregateResources field, you must include theworkloadTypefield to specify the type of workloads to run on the TPUs.

• START_TIME: the start time of the request, which youmust format as an RFC 3339 timestamp.

• END_TIME: the end time of your reservation period,which you must format as an RFC 3339 timestamp. If you want to specify aduration, in seconds, for your reservation period instead of an endtime, then replace theendTime field with theduration field.