Massed Compute review

Facts from Massed Compute’s own pricing and docs pages, September 2026 — 54 read from a quoted sentence on the page we archived.

Summary

Massed Compute rents GPUs from $0.35 per GPU-hour on demand, bills per minute, and asks no minimum commitment. Egress is free.

SOC 2 Type IIHIPAAGDPR · DPA99.999% uptime SLASpend limitsVolume discountsFree egress

From

$0.35/GPU-hr

Billing

Per minute

Minimum commitment

None

Egress

Free

Uptime SLA

99.999% claimed on product page

Registered in

🇺🇸

Visit Massed Compute ↗

On-demand pricing

NVIDIA RTX PRO 6000 Blackwell 96GB

×1, on demand · RTX PRO 6000 Blackwell (96GB)

$2.19/hr

Deploy

NVIDIA H100 SXM5 80GB

×1, on demand · H100 (80GB)

$2.73/hr

Deploy

NVIDIA L40S 48GBLowest

×1, on demand · L40S (48GB)

$0.97/hr

Deploy

NVIDIA B200 SXM6 180GB

×8, on demand · B200 SXM6

$5.43/hr

Deploy

NVIDIA B300 SXM6 288GB

×8, on demand · B300 SXM6

$6.60/hr

Deploy

See all 72 Massed Compute offers →

What’s good

Lowest published on-demand rate is $0.35/GPU-hr (NVIDIA A30 24GB), with no minimum commitment

Egress is free — no per-GB charge on data leaving

Billing is per minute

Self-serve signup — no sales call to get started

Publishes a 99.999% claimed on product page uptime commitment

Free credits available: referral credit 3% of referred spend

Plans

	On-Demand VMs	Bare Metal	GPU Clusters
From	$0.43+/hr (A30 class) Massed Compute · On-Demand VMs Every on-demand rate we hold for Massed Compute, per GPU per hour, cheapest first. NVIDIA A30 24GB ×1, on-demand · A30 (24GB) $0.35/hr Deploy NVIDIA RTX A5000 24GB ×1, on-demand · RTX A5000 (24GB) $0.44/hr Deploy NVIDIA RTX A6000 48GB ×1, on-demand · RTX A6000 (48GB) [ALT Config] $0.55/hr Deploy NVIDIA RTX 6000 Ada 48GB ×1, on-demand · RTX 6000 ADA (48GB) $0.79/hr Deploy NVIDIA L40 48GB ×1, on-demand · L40 (48GB) $0.86/hr Deploy NVIDIA RTX PRO 4500 Blackwell 32GB ×1, on-demand · RTX PRO 4500 Blackwell (32GB) $0.92/hr Deploy NVIDIA L40S 48GB ×1, on-demand · L40S (48GB) $0.97/hr Deploy NVIDIA A100 80GB ×1, on-demand · A100 (80GB) $1.35/hr Deploy NVIDIA A100 SXM4 80GB ×1, on-demand · A100 SXM4 (80GB) $1.38/hr Deploy NVIDIA RTX PRO 6000 Blackwell 96GB ×1, on-demand · RTX PRO 6000 Blackwell (96GB) $2.19/hr Deploy NVIDIA H100 SXM5 80GB ×1, on-demand · H100 (80GB) $2.73/hr Deploy NVIDIA H100 NVL 94GB ×2, on-demand · H100 NVL (94GB) NVLink $2.92/hr Deploy NVIDIA H200 NVL 141GB ×1, on-demand · H200 NVL (141GB) $3.62/hr Deploy NVIDIA B200 SXM6 180GB ×8, on-demand · B200 SXM6 $5.43/hr Deploy NVIDIA B300 SXM6 288GB ×8, on-demand · B300 SXM6 $6.60/hr Deploy Prices as last read from Massed Compute’s own pages. Open a card for every provider renting it.	Not published	Not published
Minimum term	1 hour; no contracts	—	Chosen per contract
Billing	Hourly rate, debited per minute from prepaid balance	Custom quote	Custom contract
Uptime SLA	99.999% (marketing claim, no SLA text found)	—	—
Tenancy	Virtual machines, 1× to 8× GPUs	Single-tenant physical servers, no hypervisor	InfiniBand clusters, 8–512+ nodes
Provisioning	~90 seconds	1–3 days	1–2 weeks
Payment	Credit card with auto-recharge; min balance threshold $5	—	—

Details

Billing & access

How you pay

Prepaid balance with automatic recharge

Currency

USD

Commitment discounts

Commitment pricing by contact

Free credits

Referral credit 3% of referred spend; coupon codes

Approval or KYC

Email verification, then card on file

Refunds

Per order form / refund policy (EULA)

When funds run out

Renting stops when balance falls below the minimum threshold

Late payment

14 business days to cure; persistent storage kept up to 60 days then deleted

What you get

GPUs listed

B300 SXM6, B200 SXM6, RTX PRO 6000 Blackwell, RTX PRO 4500 Blackwell, H200 NVL, H100 80GB, H100 SXM5, H100 NVL, A100 80GB, A100 SXM4, DGX A100, L40S, L40, RTX 6000 Ada, RTX A6000, RTX A5000, A30

Regions

US regions (e.g. us-central-3); locations not named

The deploy example uses a region id; no page names the data-center cities.

Node configuration

Per SKU: e.g. 1× H200 NVL 16 vCPU / 180 GB / 750 GB; 8× B200 280 vCPU / 2800 GB / 6 TB

Interconnect

InfiniBand for clusters; NVLink SKUs listed

Datacenter tier

Tier III

Local storage

Included per configuration (e.g. 725 GB with 1× RTX PRO 6000)

Persistent storage

Persistent Storage exists; pricing not on the page

Ingress

FREE (no bandwidth charges)

Images

Curated templates (PyTorch, vLLM, ComfyUI, Ollama, JupyterLab) or bring your own

Access methods

Web console, VDI, SSH, REST API, MCP server

Terraform

Terraform provider; Python, Node.js, Go SDKs

FAQ

How much does Massed Compute cost?

As of September 2026, the lowest published on-demand rate is $0.35 per GPU-hour for the NVIDIA A30 24GB. Massed Compute publishes an on-demand rate for 15 GPU models. Billing is per minute.

Does Massed Compute have an uptime SLA?

Massed Compute states 99.999% claimed on product page; no SLA document found.

Is there a minimum commitment on Massed Compute?

No. Massed Compute states none; prepaid balance with $5 minimum threshold.

Does Massed Compute charge for egress?

No. Massed Compute states that egress is free.

Can I sign up to Massed Compute without approval?

Massed Compute states that signup is self-serve. On approval and identity checks it states: email verification, then card on file.

Are Massed Compute payments refundable?

Massed Compute states: per order form / refund policy (EULA).

Where are Massed Compute's data centres?

Massed Compute lists uS regions (e.g. us-central-3); locations not named.

Who do you contract with at Massed Compute?

The contracting entity is Massed Compute Inc..

Reviewed September 2026, from Massed Compute’s own site. 54 facts, each quoted from the page it was read on.