Home / Google

Google Compute Engine GPU VMs

cloud.google.com/products/compute/pricing/accelerator-optimiGPU cloudcompute/gcp-compute-gpuJSON compare report an error provider JSON

Free allowance: none
Verdict: caveatusable, with a named limit or an unprobed read path
Cost: no price stated
Identity: email?placeholder, not evidence; card unknown, phone unknown
Binding limit: write-gatedno free-tier allowance applies to any GPU or TPU; billing starts with the first second
Evidence: documented2 sources; a first-party page was read and the sentence quoted
Freshness: stale on 2026-11-23checked 2026-10-23 (live); stale on 2026-11-23

Free tier	not published unit none; metered on wall-clock
Price	none stated
Limits	request-rate A100 80 GB from $5.0688 an hour on demand; eight-GPU H100 nodes from $93.4007 an hour source write-gated (binding) no free-tier allowance applies to any GPU or TPU; billing starts with the first second source
Identity	email (placeholder) not published
Egress	not stated
Per-object cap	not stated
Retention	not stated
Interface	unknown
Tooling	not stated
Reliability	no reliability event recorded
Evidence note	documented both the accelerator-optimized rate card and the free-features document were read on 2026-10-22. The page proves the per-GPU rate and the free-features document proves there is no free GPU allowance
Parsed numbers	request-rate 80 GB

Evidence 2

date	source	what it proved
2026-10-02	cloud.google.com/products/compute/pricing/accelerator-optimi	not published
2026-10-22	cloud.google.com/products/compute/pricing/accelerator-optimi	'GPUs and TPUs are not included in the Free Tier offer. You are always charged for GPUs and TPUs that you add to VM instances' (free-features document), against the accelerator-optimized rate card read the same day: a2-ultragpu-1g at $5.06879789 an hour on demand

More fields 10

{
 "keep_rate": null,
 "keep_rate_basis": "no field in the source",
 "billing_model": "unclassified",
 "shape": {
  "vcpu": null,
  "ram": null,
  "disk": null,
  "gpu_class": "A100 80GB",
  "gpu_count": 1,
  "note": "each accelerator-optimized type fixes its own ratio; a2-ultragpu-1g is 1 GPU, 12 vCPU, 170 GB, 275 GiB SSD"
 },
 "session_cap": {
  "wall_clock_min": null,
  "idle_reap_s": null,
  "restart_required": false,
  "note": "no free allowance and no session cap; a GPU VM bills until it is stopped or deleted"
 },
 "persistent_storage": null,
 "network_out": null,
 "price_note": null,
 "free_tier_kind": "none",
 "net": null
}

Unknown values 9

$.chaining a source was read and is silent on this field
$.egress.metered egress metering was not published
$.egress.terms a source was read and is silent on this field
$.interface the access interface was not established
$.per_object_cap.stated a source was read and is silent on this field
$.reliability_note nothing of this kind was observed
$.retention.notes the field may be stated in a source that was not read
$.retention.stated a source was read and is silent on this field
$.tooling a source was read and is silent on this field