Google Cloud Run (Compute)

cloud.google.com/run/pricingServerless runtimecompute/google-cloud-runJSON compare report an error provider JSON

Free allowance: Per billing account, monthly: 2 million requests, 180,000 vCPU-seconds and 360,000 GiB-seconds under request-based billing; 240,000 vCPU-seconds and 450,000 GiB-seconds under instance-based billing; plus 1 GiB of outbound transfer from North America
Verdict: caveatusable, with a named limit or an unprobed read path
Cost: free tier
Identity: cardcard required, no phone
Binding limit: volume-quota240,000 vCPU-seconds and 450,000 GiB-seconds a month under instance-based billing
Evidence: documented5 sources; a first-party page was read and the sentence quoted
Freshness: stale on 2026-11-25checked 2026-10-25 (live); stale on 2026-11-25

Free tier	per billing account, monthly: 2 million requests, 180,000 vCPU-seconds and 360,000 GiB-seconds under request-based billing; 240,000 vCPU-seconds and 450,000 GiB-seconds under instance-based billing; plus 1 GiB of outbound transfer from North America unit seconds; metered on wall-clock
Price	none stated
Limits	volume-quota (binding) 240,000 vCPU-seconds and 450,000 GiB-seconds a month under instance-based billing source request-rate 2 million requests a month under request-based billing source per-ip-shaping 1 GiB a month of outbound transfer from North America source retention-clamped an idle instance is billed at the idle rate when a minimum instance is configured; scale to zero is the default source it was not resolved either
Binding	volume-quota an always-on instance spends the monthly grant in about ten days: 450,000 GiB-seconds at 512 MiB is 900,000 seconds, 10.4 days.
Identity	card a card at signup; the free tier is a spending-based discount applied after the fact, not a prepaid allowance
Egress	1 GiB of data transfer within North America free per month, then Google Cloud network pricing at the Premium tier free allowance: 1 GiB
Per-object cap	a request may run up to 60 minutes; memory and CPU per instance are configurable
Retention	no persistent disk; a container is stateless unless it mounts a volume or a bucket
Interface	unknown an OCI image behind an HTTPS endpoint; the service listens on a port
Tooling	gcloud, the Cloud Run API, Docker; any language with an HTTP server
Composes with	the only free serverless tier here with 50 vCPU-hours and a listening socket
Reliability	a service scales to zero by default and pays a cold start on the next request; a minimum instance removes the cold start and starts the idle meter
Notes	Arithmetic from the published free tiers, not a measurement: at 0.25 vCPU and 512 MiB, an always-on instance needs 648,000 vCPU-seconds and 1,296,000 GiB-seconds a month. The memory grant runs out first, at 900,000 seconds, so the free tier holds an always-on 0.25 vCPU service for about 10.4 days a month. What is not verified is whether the 'Idle time (Min instance)' SKU is drawn from the same free tier; the page presents it as a separate rate.
Evidence note	documented both free-tier tables and the per-second rates were read on the pricing page
Parsed numbers	price USD 0 / month price USD 0 / month volume-quota 450000 GiB per-ip-shaping 1 GiB / month

Evidence 5

date	source	what it proved
2026-10-22	cloud.google.com/free/docs/free-cloud-features	Cloud Run free tier: 2 million requests, 360,000 GB-seconds of memory, 180,000 vCPU-seconds, 1 GB of outbound transfer a month
2026-10-24	cloud.google.com/run/pricing	quoted: 'Free tier (based on us-central1 pricing): CPU - First 240,000 vCPU-seconds free per month; RAM - First 450,000 GiB-seconds free per month' under instance-based billing
2026-10-24	cloud.google.com/run/pricing	quoted: 'CPU - First 180,000 vCPU-seconds free per month; RAM - First 360,000 GiB-seconds free per month; Requests - 2 million requests free per month' under request-based billing
2026-10-24	cloud.google.com/run/pricing	quoted: 'The free tier usage is aggregated across projects by billing account and resets every month'
2026-10-24	cloud.google.com/run/pricing	quoted rate table: 'CPU (per vCPU-second) ... $0.000018', 'Memory (per GiB-second) ... $0.000002', 'Idle time (Min instance1) $0.0000025'

More fields 17

{
 "keep_rate": {
  "value": 0.0,
  "basis": "a scaled-to-zero service does not bill",
  "source": "https://cloud.google.com/free/docs/free-cloud-features"
 },
 "keep_rate_basis": null,
 "billing_model": "per-100ms",
 "shape": {
  "vcpu": 0.25,
  "ram_gib": 0.5,
  "disk_gib": null,
  "gpu_class": null,
  "gpu_count": null
 },
 "session_cap": {
  "wall_clock_min": 60,
  "idle_reap_s": null,
  "restart_required": false,
  "note": "a request may run 60 minutes; the service scales to zero when idle unless a minimum instance is set"
 },
 "persistent_storage": {
  "included_gib": 0,
  "price_gib_month": "Cloud Storage is separate",
  "durability": "none in the runtime",
  "survives_stop": false,
  "survives_destroy": false,
  "max_retention": "none"
 },
 "network_out": {
  "allow_outbound": true,
  "allow_inbound_listen": true,
  "ports": "the container port on a generated HTTPS hostname",
  "static_ip": false,
  "ipv6": false,
  "per_session_bytes": null
 },
 "price_note": null,
 "granularity": {
  "min_billed_s": 0.1,
  "rounding_s": 0.1
 },
 "free_tier_kind": "monthly-spending-discount",
 "open_question": "whether the grant applies to the Services meter or the Instances meter",
 "gap": "the grant's meter (Services or Instances) is not resolved; the claim page's Instances section and footnotes would close it",
 "burst_x": 1.0,
 "term": "monthly",
 "m10h_usd": 0.0,
 "m24h_usd": 0.0,
 "price_arithmetic": "free within the monthly grant; 512 MiB at 0.25 vCPU consumes the 450,000 GiB-second grant in 900,000 s = 10.4 days of 24/7"
}

Unknown values 1

$.interface the access interface was not established