v1

latestOpenAPI 3.1.02026-07-13897404.3 KB

search offers

Search for available GPU machine offers with advanced filtering and sorting.

Each filter parameter (such as verified, gpu_name, num_gpus, etc.) should be an object specifying the operator and value you want to match.

Filter operators:

Operator	Meaning	Example
eq	Equal to	{ "eq": true }
neq	Not equal to	{ "neq": false }
gt	Greater than	{ "gt": 0.99 }
lt	Less than	{ "lt": 10000 }
gte	Greater than or equal	{ "gte": 4 }
lte	Less than or equal	{ "lte": 8 }
in	Value is in a list	{ "in": ["RTX_3090", "RTX_4090"] }
notin	Value is not in a list	{ "notin": ["TW", "SE"] }

CLI Usage: vastai search offers 'reliability > 0.99 num_gpus>=4' --order=dph_total

post/api/v0/bundles/

Request body

limitinteger

Max offers to return

type'ondemand' | 'bid' | 'reserved'

Instance type for the offer. Affects pricing calculation.

ondemand: Fixed pricing based on listed rates. Default.
bid (interruptible): Uses minimum bid price. Lower cost but may be interrupted if outbid.
reserved: Reserved instance pricing.

reliabilityobject

Machine reliability score (0-1). Example: {"gte": 0.99}

gpu_ramobject

GPU RAM in MB. Example: {"gte": 24000}

durationobject

Minimum required rental duration in seconds (the offer must be available for at least this long from now).

machine_idobject

Filter by specific host machine ID

dlperf_per_dphtotalobject

DLPerf per dollar per hour

dph_totalobject

Total $/hour rental cost. Example: {"lte": 0.5}

flops_per_dphtotalobject

TFLOPs per $/hour

gpu_archobject

Host machine GPU architecture (e.g. nvidia, amd). Example: {"eq": "nvidia"}

dlperfobject

Deep Learning performance score

cuda_max_goodobject

Maximum supported CUDA version

inet_downobject

Download bandwidth (MB/s)

inet_upobject

Upload bandwidth (MB/s)

inet_down_costobject

Download bandwidth cost ($/GB)

inet_up_costobject

Upload bandwidth cost ($/GB)

driver_versionobject

NVIDIA driver version in the format "XXX.XX.XX"

compute_capobject

CUDA compute capability x 100. Use 650 for compute capability 6.5 or 700 for 7.0.

cpu_archobject

Host machine CPU architecture (Default is amd64). Example: {"eq": "amd64"}

has_avxobject

CPU supports AVX instruction set

cpu_coresobject

Number of virtual CPUs

cpu_cores_effectiveobject

Effective vCPU count for the offer

cpu_ghzobject

CPU clock speed in GHz

cpu_ramobject

CPU RAM in MB

datacenterobject

Show only datacenter offers

externalobject

Show external offers in addition to datacenter offers

disk_bwobject

Disk read bandwidth in MB/s

disk_spaceobject

Disk storage space in GB

bw_nvlinkobject

NVLink interconnect bandwidth in GB/s. Use when filtering for multi-GPU systems with high-speed NVLink.

gpu_max_powerobject

GPU power limit in watts

gpu_max_tempobject

GPU temperature limit in Celsius

gpu_mem_bwobject

GPU memory bandwidth in GB/s

gpu_total_ramobject

Total GPU RAM across all GPUs in MB

gpu_fracobject

Fraction of the total GPU resources being offered

gpu_display_activeobject

Whether the GPU has an attached display

direct_port_countobject

Number of direct ports

host_idobject

Host user ID

idobject

Offer ID

min_bidobject

Minimum bid price ($/hour)

mobo_nameobject

Motherboard name

pci_genobject

PCIe generation

pcie_bwobject

PCIe bandwidth (CPU to GPU)

storage_costobject

Storage cost in $/GB/month

static_ipobject

Whether the IP address is static/stable

total_flopsobject

Total theoretical GPU compute performance (TFLOPs) across all GPUs.

os_versionobject

Host machine Ubuntu OS version

ubuntu_versionobject

Host machine Ubuntu OS version (alias for os_version)

verificationobject

Machine verification status string (verified, deverified, unverified). Example: {"eq": "verified"}

vms_enabledobject

Whether the machine is a VM instance

allocated_storagenumber

Storage allocation size in GB for the instance. This sets the disk size when creating the instance and cannot be changed later. Default is 8GB.

Example request

{
  "gpu_name": {
    "in": [
      "RTX_3090",
      "RTX_4090"
    ]
  },
  "num_gpus": {
    "in": [
      1,
      2,
      4,
      8
    ]
  },
  "geolocation": {
    "in": [
      "US",
      "CA"
    ]
  }
}

Response

Successful search response

offersobject[]