Compute Comparison

Compare listed Oracle Cloud GPU hourly rates, models and regions below. Check the provider for availability, full instance cost and billing terms before choosing.

Oracle Cloud

Hyperscaler

GB200 NVL72 superclusters up to 131,072 GPUs — NVIDIA and AMD GPU catalog across hyperscale and bare-metal

Oracle Cloud Infrastructure (OCI) offers a wide GPU catalog across hyperscaler and bare-metal tiers: NVIDIA GB200 Grace Blackwell Superchips in superclusters scaling to 131,072 GPUs, HGX B200, H200, H100, L40S, A100, A10, and AMD Instinct MI355X, MI300X, and MI325X. Bare-metal instances deliver full hardware performance with no virtualization overhead. RDMA cluster networking provides up to 3,200 Gbps bandwidth. NVIDIA DGX Cloud on OCI provides serverless AI training. OCI Supercluster delivers 20 exaflops aggregate. AMD MI355X available via try-before-you-buy program.

Visit Oracle Cloud

Cheapest On-Demand Oracle Cloud Pricing

$2.00/hr

Cheapest Spot Oracle Cloud Pricing

—

Oracle Cloud GPU Listings

3

Oracle Cloud Billing

Pay-as-you-go, Annual Flex, Try-before-you-buy (MI355X)

Oracle Cloud Performance Benchmarks

Loading…

Compare Oracle Cloud With Another Cloud Provider

Oracle Cloud Info

Headquarters

Austin, TX

Founded

2016

Regions

us-ashburn-1, us-phoenix-1, eu-frankfurt-1, ap-tokyo-1, ap-sydney-1

Min Commitment

None (pay-as-you-go)

Support

Basic → Premier

Oracle Cloud Strengths

  • ▸GB200 NVL72 superclusters scaling to 131,072 GPUs
  • ▸AMD MI355X — 192 GB VRAM for large-model inference
  • ▸Bare-metal instances — no virtualization overhead
  • ▸NVIDIA DGX Cloud on OCI for serverless AI training
  • ▸3,200 Gbps RDMA cluster networking
  • ▸AMD MI300X and MI325X available alongside NVIDIA catalog

Oracle Cloud Weaknesses

  • ▸Smaller ecosystem and fewer ML tooling integrations vs AWS/GCP
  • ▸Support quality inconsistent outside enterprise tiers
  • ▸AMD GPU availability limited vs NVIDIA options

Oracle Cloud is Best For

Hyperscale AI training on GB200 superclustersLarge-model inference on AMD MI300X/MI355X (192 GB VRAM)Oracle database workloads with AICost-sensitive enterprise teams comparing hyperscaler pricing

Oracle Cloud GPU Catalog

← swipe to scroll →
GPU ModelvRAMOn-DemandSpotAvailabilityRegionSource
A1024 GB$2.00—Medus-ashburn-1live APIretrieved 9m ago
A100 80GB80 GB$4.00—Medus-ashburn-1live APIretrieved 9m ago
H100 80GBoutlier80 GB$10.00—Medus-ashburn-1live APIretrieved 9m ago

Oracle Cloud Reviews

No reviews yet

Your experience could help other engineers choosing between providers.

Oracle Cloud — Frequently Asked Questions (FAQs)

Browse Other GPU Providers

Oracle Cloud GPU pricing overview

Oracle Cloud is a hyperscaler GPU cloud provider headquartered in Austin, TX. Billing is Pay-as-you-go, Annual Flex, Try-before-you-buy (MI355X) with a minimum commitment of None (pay-as-you-go). Available regions include us-ashburn-1, us-phoenix-1, eu-frankfurt-1, ap-tokyo-1 and 1 more. GPU instances can be provisioned on-demand, making Oracle Cloud suitable for both short-duration experiments and sustained production workloads.

Oracle Cloud vs other GPU providers

Oracle Cloud competes with providers including Lambda Labs, CoreWeave, RunPod, Paperspace, Vast.ai, and the major hyperscalers (AWS, Google Cloud, Azure) for GPU compute workloads spanning LLM training, fine-tuning, and inference serving. Key differentiators include: GB200 NVL72 superclusters scaling to 131,072 GPUs; AMD MI355X — 192 GB VRAM for large-model inference; Bare-metal instances — no virtualization overhead. Use the side-by-side comparison tool above to see Oracle Cloud pricing against any other provider across shared GPU models. For a broader market view, the live GPU prices table shows all 3 Oracle Cloud listings alongside 91+ providers in a single sortable view.

Best use cases for Oracle Cloud

Oracle Cloud is best suited for: Hyperscale AI training on GB200 superclusters, Large-model inference on AMD MI300X/MI355X (192 GB VRAM), Oracle database workloads with AI, Cost-sensitive enterprise teams comparing hyperscaler pricing. Support tiers range from Basic → Premier, making it viable for both individual researchers and enterprise teams with SLA requirements. There are currently 3 active GPU listings on Oracle Cloud, covering A100 80GB, A10, H100 80GB. H100 SXM5 instances with NVLink interconnect are commonly used for large-scale training workloads. A100 40GB and RTX 4090 instances are frequently used for fine-tuning and inference of smaller models.

Oracle Cloud billing model and cost structure

Oracle Cloud uses Pay-as-you-go, Annual Flex, Try-before-you-buy (MI355X) pricing. Spot pricing is not currently available on this provider — all instances are on-demand. Reserved instance pricing may be available for predictable long-running workloads — check the provider's pricing page for current committed-use rates. Always compare the effective hourly rate including egress, storage, and networking costs when evaluating total cost of ownership across providers.

Choosing the right GPU on Oracle Cloud

GPU selection depends on model size, precision, and whether your workload is compute-bound or memory-bandwidth-bound. For LLM training above 30B parameters, H100 80GB SXM5 instances with NVLink are the standard choice — high HBM3 bandwidth and FP16 throughput make them well-suited for large transformer workloads. For inference of 7B–13B models in FP16 or BF16, A100 40GB is a common choice on many providers. RTX 4090 instances are widely used for fine-tuning, prototyping, and quantized inference (INT4/INT8) of models up to 70B. Read the H100 vs A100 guide or the GPU benchmarks for ML guide for a full breakdown.

How Oracle Cloud pricing data is collected

Prices shown are sourced from Oracle Cloud's public pricing API or pricing page and refreshed every 15 minutes. On-demand rates reflect the current list price for a single GPU instance in the cheapest available region. Spot prices, where available, reflect interruptible instance rates at the time of the last snapshot. All prices are in USD per hour. Daily snapshots are kept in our history archive and visualised in the GPU price history charts — useful for identifying seasonal pricing patterns and evaluating whether current rates are above or below the 30-day average.

Evaluating managed LLM inference APIs as an alternative to self-hosted GPU compute? Compare live LLM token prices across OpenAI, Anthropic, Google, Groq, and many other providers. The cheapest GPU cloud guide covers the break-even analysis between self-hosted and managed inference at different request volumes.

Ready to get started?

Launch your first GPU on Oracle Cloud

On-demand from $2.00/hr — 3 GPU configurations available. Pay-as-you-go, Annual Flex, Try-before-you-buy (MI355X) billing.

Go to Oracle Cloud
Oracle Cloud
Visit Oracle Cloud