← All Providers

GCP GPU prices

Track GCP GPU prices for H100 SXM, H200, A100 SXM4, and B200, compare current median rates, and see how this provider moves versus the broader cloud market.

Plan with GCP Browse GPU prices Open GCP
Lowest Median
—
GPUs Available
—
models tracked
Median Range
—
across recent GPUs
Last Updated
—
Search guide

GCP cloud GPU pricing overview

GCP currently appears in our dataset with pricing for 7 tracked GPU models, including H100 SXM, H200, A100 SXM4, and B200.

Use this page when you want a provider-first view: start with the current table to compare median rates today, then use the history chart to understand how the provider's market has moved over time.

Cheapest provider right now

Lowest tracked GCP GPU median right now

The lowest tracked GCP GPU median right now is L4 at $0.11/hr (spot).

Methodology and freshness

How provider pricing is normalized

We normalize each tracked listing into per-GPU hourly pricing, keep the latest snapshot per pricing type, and let you compare the same provider across many GPU models without manually reconciling marketplace formats.

FAQ

GCP GPU pricing FAQ

Does GCP offer H100, A100, or H200 GPUs?

GCP currently has tracked pricing for H100 SXM, H200, A100 SXM4, and B200.

What is the lowest tracked GPU median on GCP right now?

The lowest tracked GCP GPU median right now is L4 at $0.11/hr (spot).

How fresh are the GCP GPU prices on this page?

This page shows the latest stored snapshot for each GCP GPU and pricing type. The freshest row currently visible is from Jul 31, 2026, and collectors run on a daily cadence.

How should I compare GCP with other GPU providers?

Start with the current GPU table to compare median prices on this provider, then use the provider comparison links below to compare shared GPUs against hyperscalers and specialist GPU clouds.

Popular follow-up searches

Popular GCP GPU price searches

These follow-up pages prioritize the GPU and provider combinations most likely to matter for buyers researching rental cost, availability, and alternatives.

GCP H100 SXM price Compare GCP's current H100 SXM rate against the rest of the tracked market. GCP H200 price Compare GCP's current H200 rate against the rest of the tracked market. GCP A100 SXM4 price Compare GCP's current A100 SXM4 rate against the rest of the tracked market. GCP B200 price Compare GCP's current B200 rate against the rest of the tracked market. Cheapest GPU for GLM-4.7-Flash Map the 3B-active GLM-4.7-Flash onto the current 32GB and 48GB GPU market. Best GPU for Kimi Linear 48B hosting See which 48GB-class GPU is the cheapest practical home for Kimi's long-context 3B-active model.
Coverage notes

How GCP pricing is tracked

GCP rows are most useful for teams that already keep data, networking, or ML tooling on Google Cloud.

Source
Google Cloud Billing API Requires `GCP_API_KEY`
Pricing models
on-demand, spot tracked where available
Best for
GCP-native ML stacks, enterprise baselines, and integrations with adjacent Google Cloud services. buyer fit
Region support: Region metadata is preserved when returned by the billing API.
Limitations: Committed-use SKUs are intentionally excluded; reserved and Dynamic Workload Scheduler SKUs are excluded too. Collection also depends on API-key health, quota, and billing catalog shape changes.
Workload views

GCP prices by workload

Jump from the full provider catalog into workload-specific slices for model serving, batch queues, fine-tuning, and training.

GCP vLLM hosting prices OpenAI-compatible vLLM serving usually starts at 24GB GPUs and moves into 80GB+ rows for larger models, batching, and long prompts. GCP batch inference prices Batch inference can use cheaper and less stable pricing modes when queues are flexible and retries are acceptable. GCP fine-tuning prices Fine-tuning usually starts in 80GB-class inventory once optimizer state, activations, checkpointing, and sequence length are considered. GCP training prices Training budgets are dominated by 80GB+ accelerators, provider stability, and whether newer H100/H200/B200/MI300X rows justify their premium over A100.

Current GCP GPU prices

Loading freshness status...

GPU VRAM Type Min Median Max Offers Last Updated
Loading...

Compare GCP against other providers

GCP vs Vast.ai GCP vs AWS GCP vs Azure GCP vs Lambda GCP vs RunPod GCP vs Oracle
Agent-assisted setup

Hand this runbook to Claude Code or Codex

Open a terminal in the repository where you want the deployment files, start claude or codex, then paste this prompt. It asks the agent to verify sources and stop before it creates billable infrastructure.

Claude Code quickstart ↗ Codex docs ↗
Deployment prompt 
Use the infrastructure or model context on this page to create a reproducible open-model deployment. Use this guide as the starting context: https://www.getflops.ai/provider/gcp. Read the linked model card and provider documentation before choosing hardware or runtime settings. Open every linked primary source and flag any mismatch instead of guessing. Create a deployment folder containing README.md, .env.example with no secrets, a pinned start script or infrastructure manifest, and smoke-test.sh. Make the endpoint OpenAI-compatible where the runtime supports it. Run local/static validation, estimate the billable resources, and stop before provisioning paid infrastructure until I approve.
Guardrails included No secrets in files · verify primary docs · approval before spend