Cloud pricing, independently compared
Find the cheapest cloud for your workload
Real per-hour and per-GB numbers for GPU cloud, inference APIs, servers, object storage, databases, Kubernetes, CDN, and egress - on one page, with a transparent methodology.
Not sure where to start? Take the 60-second cloud finder quiz
Browse pricing
Every layer of your stack, priced
Pick a category and compare real, normalized pricing across every provider that offers it.
GPU cloud
H100, A100, B200 and more, by the hour.
ExploreServerless GPU
Serverless inference at scale, pay-per-second.
ExploreLLM inference
Token pricing across hosted model APIs.
ExploreVPS / cloud servers
Always-on compute from budget to bare metal.
ExploreBare metal
Dedicated physical servers, hourly or monthly.
ExploreServerless CPU / functions
Pay for execution time, not idle capacity.
ExploreObject storage
S3-compatible buckets and egress fees.
ExploreBlock storage
Persistent disk volumes with provisioned IOPS.
ExploreManaged databases
Postgres, MySQL and more, fully managed.
ExploreKubernetes
Control-plane and node pricing, compared.
ExploreCDN
Edge delivery and per-GB request costs.
ExploreEgress
What it really costs to move data out.
ExploreGPU pricing
Cheapest GPU cloud right now
On-demand per-hour pricing, normalized across providers.
| GPU | Lowest price | |
|---|---|---|
| NVIDIA GB200 | $5.50 /GPU-hr | Compare |
| NVIDIA B200 | $3.50 /GPU-hr | Compare |
| AMD Instinct MI300X | $1.99 /GPU-hr | Compare |
| NVIDIA H200 | $2.70 /GPU-hr | Compare |
| NVIDIA H100 | $1.95 /GPU-hr | Compare |
| NVIDIA A100 80GB | $1.15 /GPU-hr | Compare |
Providers
Featured cloud providers
The providers we track across every category - from hyperscalers to niche GPU clouds.
Hetzner
Germany
Amazon Web Services
United States
Anthropic
United States
Cloudflare
United States
OpenAI
United States
Backblaze B2
United States
Bunny.net
Slovenia
DigitalOcean
United States
Why DeployCue
Built for engineers who comparison-shop
Numbers, not marketing
Per-hour and per-GB list prices pulled straight from provider pricing pages and APIs. Normalized so they actually compare. Some links are affiliate links, but rankings are strictly price-driven.
Spot, reserved & egress
We surface the costs that bite later - spot instance rates, egress per GB, and committed-use discounts. Not just the headline on-demand price.
Independent & transparent
Some links are affiliate links, but rankings are strictly price-driven. Our methodology is documented, reproducible, and open to scrutiny.
Learn
Cloud guides & pricing insight
Deep dives into cloud pricing, provider comparisons, and infrastructure decisions.
Deploying Mixtral and MoE Models: Cost Quirks of Sparse Experts
Mixture-of-experts models like Mixtral are cheap to run but expensive to hold in memory. That quirk drives every cost decision.
Inference Autoscaling: Handling Traffic Spikes Without Overpaying
Autoscaling inference well means absorbing spikes without paying for idle GPUs the rest of the time. Here is how to tune it.
Continuous Batching: The Trick Behind High-Throughput LLM Serving
Continuous batching keeps the GPU busy by swapping finished requests for new ones mid-flight. It is why modern serving is so efficient.
GPU Sizing for LLM Serving: Matching VRAM to Model Size
Pick a GPU too small and the model will not load; too big and you overpay. Here is how to size VRAM to your model.
Head to head
Compare providers side by side
See how providers stack up on pricing, features, and regions. Data-driven, no bias.
Calculators