# GPUBank vs Baseten

> Baseten is a San Francisco inference platform, founded in 2019, that runs models as autoscaled dedicated replicas billed per minute, plus training jobs and per-token Model APIs. It is not a raw VM rental. Its H100 costs $0.11 per minute, or $6.50 per GPU-hour; a GPUBank H100 SXM costs $1.85. Baseten sells autoscaling, scale-to-zero, a 99.9 % SLA and compliance reports; GPUBank sells plain GPU instances with SSH, billed by the second.

- Page: https://gpubank.io/compare/gpubank-vs-baseten
- Prices last checked on 29 September 2026, on Baseten's own pages (sources below). Page updated 2 November 2026.
- Baseten: https://www.baseten.co, headquartered in San Francisco, United States
- GPUBank: https://gpubank.io, GPU cloud billed per second at an all-in hourly price (vCPU, RAM, NVMe, one IPv4 and traffic included), prepaid balance topped up in crypto.

## Verdict

**Managed inference and compliance at Baseten. Raw GPUs at a fraction of the price at GPUBank.**

Baseten fits teams that serve models in production and want autoscaling, scale-to-zero, managed inference engines, a contractual SLA and SOC 2, ISO 27001 or HIPAA reports. You deploy a model package or container, not a machine. For the four GPUs both of us offer, GPUBank costs 58 to 72 % less per GPU-hour, and it bills by the second, not per started minute with cold starts and image builds on the bill. GPUBank gives a full instance with SSH, JupyterLab and persistent disks, paid in crypto; Baseten bills a card or US bank account after the fact.

Score over 26 points: GPUBank ahead on 12, even on 5, Baseten ahead on 9.

## Key facts

| Criterion | Baseten | GPUBank |
|---|---|---|
| H100 SXM, per GPU-hour | $6.50 ($0.11 per minute) | $1.85, all-in |
| Cheapest GPU-hour | $0.63, T4 | $0.23, RTX 4090 |
| Billing granularity | Per minute, partial minutes rounded up | Per second |
| Stopped instance | Scaled to zero: no charge | Disk only |
| Persistent storage | Not published | $0.03 per GB-month |
| Egress traffic | Not published | Included |
| Crypto payments | Not listed; card or US bank account | Bitcoin, USDT, Ethereum and 12+ more |
| Card payments | Cards and US bank accounts | No |
| Minimum deposit | None, postpaid | $25 |
| GPU models listed | 9, including a 40 GB H100 MIG slice | 10 |
| Regions | 2 region slugs, us and eu | 4 |
| Serverless GPUs | Autoscaled replicas, scale to zero | No |
| Availability target | 99.9 % monthly, credits up to 40 % | 99.9 % monthly |
| Compliance reports | SOC 2 Type II, ISO 27001, HIPAA | None published |
| Support | Email and chat; Slack and Zoom on Pro | Tickets from a $25 balance |
| Headquarters | San Francisco, United States | Tallinn, Estonia |

## Same GPUs, price per GPU-hour (USD, excl. VAT)

| GPU | Baseten | GPUBank | Difference |
|---|---|---|---|
| B200 SXM | $9.98 (on demand) | $4.15 | -58 % |
| H200 SXM | $7.50 (on demand) | $2.49 | -67 % |
| H100 SXM | $6.50 (on demand) | $1.85 | -72 % |
| RTX PRO 6000 | $4.00 (on demand) | $1.15 | -71 % |

## When to pick Baseten

- Autoscaled model replicas that scale to zero when idle
- 99.9 % monthly SLA with service credits of 10 to 30 %
- SOC 2 Type II, ISO 27001, HIPAA and PCI DSS listed
- Per-token Model APIs and multi-node InfiniBand deployments
- Postpaid invoices, no deposit needed

## When to pick GPUBank

- H100 SXM at $1.85 per GPU-hour, against $6.50
- Per-second billing that starts after the health check
- A full GPU instance with SSH, JupyterLab and persistent disks
- Four named regions to pick from at launch
- Pay in Bitcoin, USDT and 12+ other coins, no card

## Questions

### Is GPUBank cheaper than Baseten for an H100?

Yes. Baseten lists its H100 at $0.11 per minute, which is $6.50 per GPU-hour. GPUBank charges $1.85 per GPU-hour for an H100 SXM, with vCPU, RAM, NVMe, IPv4 and traffic included. The two products differ: Baseten runs autoscaled model replicas, GPUBank rents GPU instances.

### Is Baseten a GPU cloud or an inference platform?

An inference platform. You deploy a model or container as dedicated replicas that autoscale and can scale to zero, or call per-token Model APIs. There is no general-purpose VM or Jupyter product, and SSH into inference containers must be enabled by support.

### Do idle Baseten replicas cost money?

Yes, unless the deployment scales to zero. Warm replicas with min_replica 1 or more are billed, and the default scale-down delay of 900 seconds keeps a replica billed for 15 minutes after traffic stops. Cold starts and image builds are billed too.

### Does Baseten accept crypto?

Baseten lists card or US bank account through its payment processor, billed after use. No crypto or PayPal option is published. GPUBank is paid from a USD balance topped up in Bitcoin, USDT, Ethereum and more than a dozen other coins.

### Which has the stronger SLA and compliance?

Baseten. It publishes a 99.9 % monthly SLA with service credits of 10, 15 or 30 % of the bill, and lists SOC 2 Type II, ISO 27001, HIPAA and PCI DSS on its Trust Center. GPUBank targets 99.9 % monthly per region in its terms and publishes no SOC 2 or ISO 27001 report.

### Are these prices with or without VAT?

Without. Every price on this page excludes VAT and sales tax, for Baseten and for GPUBank, so the two columns compare like for like. Tax is added at checkout according to where the buyer is based.

## Sources

- https://www.baseten.co/pricing/: Plans, per-minute GPU prices on the page, Model API prices, FAQ (credits, idle time, support, discounts)
- https://docs.baseten.co/deployment/resources: Full GPU instance table with per-minute prices, vCPU, RAM, VRAM, GPU specs
- https://docs.baseten.co/organization/billing: Per-minute metering, lifecycle billing, payment methods, credits, invoice cadence, x60 conversion
- https://docs.baseten.co/deployment/autoscaling/overview: Scale-to-zero, default scale-down delay 900 s
- https://docs.baseten.co/deployment/regional-deployments: Regions us and eu, verified-organization requirement
- https://docs.baseten.co/hosting-options/cloud: Multi-cloud pool, GPUs never shared, self-service
- https://docs.baseten.co/hosting-options/single-tenant: Enterprise custom pricing, private networking
- https://docs.baseten.co/observability/security: Data retention, isolation, pentesting
- https://docs.baseten.co/reference/overview: APIs, CLIs, SDKs
- https://docs.baseten.co/inference/ssh: SSH into deployments, enabled on request
- https://docs.baseten.co/training/concepts/cache: Training storage behaviour
- https://docs.baseten.co/inference/model-apis/pricing-and-limits: Account tiers Basic unverified/verified, rate limits
- https://www.baseten.co/service-level-agreement/: 99.9% SLA and service credits
- https://status.baseten.co/: 90-day uptimes
- https://trust.baseten.co: SOC 2, ISO 27001, HIPAA, PCI DSS, FedRAMP 20x, subprocessors
- https://www.baseten.co/terms-and-conditions/: Legal name, San Francisco address, invoice terms
- https://www.baseten.co/about-us/: Founded 2019, founders
- https://www.baseten.co/deployments/baseten-cloud/: Deployment options, compliance claims

Prices change: check both providers before buying. Baseten is a trademark of its owner; GPUBank is not affiliated with Baseten.