SAVRN Index · GPU cloud
Impossible Cloud GPU Pricing and Company Profile
Privately held
impossiblecloud.com ↗
Impossible Cloud is a European cloud provider headquartered in Hamburg, Germany, founded in 2021 by Dr. Kai Wawrzinek and Dr. Christian Kaul. It sells S3-compatible object storage, dedicated NVIDIA bare-metal GPU servers (B300, H200 and H100) in EU, US and APAC regions, and an OpenAI-compatible Serverless Inference API. Its GPU servers are priced per GPU-hour on committed terms of 6 to 60 months, published in a public JSON feed that the SAVRN Index reads directly. The company says it is ISO/IEC 27001 certified and serves more than 3,000 enterprise customers.
Prices tracked
12
12 confirmed on the provider's feed or page at the last check
Sources
2
Every cited page, listed below
Market
Impossible Cloud is privately held, so there is no public market data to show.
Against the market
How Impossible Cloud prices compare
Impossible Cloud's lowest price for each GPU beside the lowest price every other tracked provider lists for the same GPU, sold the same way. Typical is the middle provider, one price each.
NVIDIA H100Reserved
All NVIDIA H100 prices →Impossible Cloud$2.95per GPU-hour, 12-month term
Rank2 of 914% below typical
Typical$3.42middle of 9 providers
Lowest anywhere$2.30Vultr
Vultr1. NVIDIA HGX H100 — Bare Metal (8 × H100 80 GB SXM)
$2.30
Impossible Cloud2. H100 SXM 8-GPU nodes (EU, offer 286-H100)
$2.95
Together AI3. reserved
$3.19
DigitalOcean4. 12-month reserved
$3.26
Cirrascale5. 8X NVIDIA H100 (Standalone)
$3.42
Astra Intellectus6. H100 SXM 8-GPU
$3.65
Microsoft Azure7. Standard_ND96is_H100_v5 (no InfiniBand variants incl. _noIB_, _flex_)
$4.86
Google Cloud8. 3-year committed use
$4.86
Lambda9. 1-Click Clusters — NVIDIA H100
$5.54
Impossible CloudTypical $3.42
NVIDIA H200Reserved
All NVIDIA H200 prices →Impossible Cloud$3.11per GPU-hour, 12-month term
Rank1 of 815% below typical
Typical$3.64middle of 8 providers
Lowest anywhere$3.11Impossible Cloud
Impossible Cloud1. H200 SXM 8-GPU nodes (EU, offer 287-H200)
$3.11
TheAI Cloud2. 1-year reserved
$3.20
DigitalOcean3. 12-month reserved
$3.40
Cirrascale4. 8X NVIDIA H200
$3.63
Astra Intellectus5. H200 SXM 8-GPU
$3.65
Together AI6. reserved
$3.99
Google Cloud7. 3-year committed use
$4.65
Microsoft Azure8. Standard_ND96isr_H200_v5 (ND H200 v5)
$5.28
Impossible CloudTypical $3.64
NVIDIA B300Reserved
All NVIDIA B300 prices →Impossible Cloud$4.32per GPU-hour, 60-month term
Rank2 of 410% below typical
Typical$4.81middle of 4 providers
Lowest anywhere$3.65Astra Intellectus
Astra Intellectus1. B300 SXM 8-GPU (US #1, 128 nodes)
$3.65
Impossible Cloud2. B300 SXM 8-GPU nodes (US, offer 290-B300)
$4.32
TheAI Cloud3. 24-month reserved
$5.30
DigitalOcean4. NVIDIA HGX B300
$7.94
Impossible CloudTypical $4.81
By contract length
What a longer commitment buys
NVIDIA B3004 contract lengths
Committing 5 years instead of 1 year cuts the B300 price 45%, from $7.89 to $4.32 per GPU-hour: $25,229 a month for one 8-GPU server instead of $46,078.
1 yearB300 SXM 8-GPU nodes (EU, offer 288-B300)
$7.89
Shortest$46,078 a month per 8-GPU server
2 yearsB300 SXM 8-GPU nodes (EU, offer 293-B300)
$6.53
17% less$38,135 a month per 8-GPU server
3 yearsB300 SXM 8-GPU nodes (EU, offer 293-B300)
$4.74to $5.47
40% less$27,682 a month per 8-GPU server
5 yearsB300 SXM 8-GPU nodes (US, offer 290-B300)
$4.32to $4.84
45% less$25,229 a month per 8-GPU server
Tracked prices
Impossible Cloud prices
Every price the Index tracks for Impossible Cloud, in the provider's own words beside a normalized figure, with the date it was last checked. What AI Costs compares them across the market.
| GPU | Offer | Terms | Published price | Per GPU-hour | Change | Checked | Source |
|---|---|---|---|---|---|---|---|
| NVIDIA H100 | H100 SXM 8-GPU nodes (EU, offer 286-H100)EU | 12-month termCommitted | $2.95 per GPU-hour, 12-month term, 12 nodes x 8 GPUs (96 GPUs), EU | $2.95 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA H200 | H200 SXM 8-GPU nodes (EU, offer 287-H200)EU | 12-month termCommitted | $3.11 per GPU-hour, 12-month term, 2 nodes x 8 GPUs (16 GPUs), EU | $3.11 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA H200 | H200 SXM 8-GPU nodes (APAC, offer 294-H200)APAC | 6-month termCommitted | $3.16 per GPU-hour, 6-month term, 8 nodes x 8 GPUs (64 GPUs), APAC, available 8 Oct 2026 | $3.16 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (US, offer 290-B300)US | 60-month termCommitted | $4.32 per GPU-hour, 60-month term, 4 nodes x 8 GPUs (32 GPUs), US | $4.32 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (EU, offer 293-B300)EU | 36-month termCommitted | $4.74 per GPU-hour, 36-month term, 1 node x 8 GPUs (8 GPUs), EU, available 1 Jan 2027 | $4.74 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 8-GPU nodes (US, offer 289-B300)US | 60-month termCommitted | $4.84 per GPU-hour, 60-month term, 16 nodes x 8 GPUs (128 GPUs), US, available 2026 Dec (RFS), 8-12 weeks after MSA | $4.84 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (EU, offer 291-B300)EU | 36 to 60-month termCommitted | $4.95 per GPU-hour, 36 to 60-month term, 64 nodes x 8 GPUs (512 GPUs), EU, available 1 Mar 2027 | $4.95 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (US, offer 292-B300)US | 36-month termCommitted | $5.21 per GPU-hour, 36-month term, 24 nodes x 8 GPUs (192 GPUs), US | $5.21 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (US, offer 290-B300)US | 36-month termCommitted | $5.37 per GPU-hour, 36-month term, 4 nodes x 8 GPUs (32 GPUs), US | $5.37 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 8-GPU nodes (US, offer 289-B300)US | 36-month termCommitted | $5.47 per GPU-hour, 36-month term, 16 nodes x 8 GPUs (128 GPUs), US, available 2026 Dec (RFS), 8-12 weeks after MSA | $5.47 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (EU, offer 293-B300)EU | 24-month termCommitted | $6.53 per GPU-hour, 24-month term, 1 node x 8 GPUs (8 GPUs), EU, available 1 Jan 2027 | $6.53 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
| NVIDIA B300 | B300 SXM 8-GPU nodes (EU, offer 288-B300)EU | 12-month termCommitted | $7.89 per GPU-hour, 12-month term, 60 nodes x 8 GPUs (480 GPUs), EU | $7.89 | No change | FeedOct 9, 2026 | console.impossiblecloud.com ↗ |
Sources
Sources for Impossible Cloud
Every source cited on this page, once each.
- impossiblecloud.comImpossible Cloud homepage and FAQ
- console.impossiblecloud.comImpossible Cloud GPU offers feed
GPU cloud
More in the Index
Astra IntellectusBasetenCirrascaleDellDenvr DataworksDigitalOceanFireworks AI (dedicated)GMI CloudHPEHyperbolicLightning AIMassed ComputeRunpodTensorWaveTheAI CloudVast.ai (marketplace medians)Verda
Capital figures come from the Capital Atlas, refreshed every Tuesday, Thursday and Saturday. Site commitments overlap reported capital expenditure and overlap one another, so each figure appears as its source states it and none are added into a single deployment number. Informational, not investment advice.
Sources
The SAVRN reference set
Seven references, one system
Capital, delay, policy, water, grid, permits and scarcity are one system seen from seven sides. Each reference is primary-sourced, refreshed on a schedule, and links into the others, so a project can be followed from the money committed to the docket that stopped it.
Capital Atlas
Who is funding AI data centers: capital providers, named sites, financing transactions, and the blocked-and-delayed record, state by state and company by company.
Open the atlas →
Delay Watchlist
Project-level slippage: named projects, jurisdictional actions, utility and federal dockets, SEC disclosures, and the interconnection queue.
Open the watchlist →
Moratorium Tracker
The policy layer: every state measure and local moratorium mapped and cited, with one page per state.
Open the tracker →
Water Tracker
Every federal, state, local and interstate water measure that touches a data center, plus a fifty-state water ledger.
Open the water tracker →
Grid Operator Watchlist
ERCOT, PJM, MISO, CAISO, SPP, NYISO, ISO-NE and FERC: the large-load rules, filings and queue positions that decide whether a site can energize.
Open the watchlist →
Permits and Power Development
Data center and substation permits, large loads in the interconnection queues, and the power plants planned to serve them, refreshed hourly from public records.
Open the permits tracker →
SAVRN Scarcity Tracker
Eleven primary-sourced indicators of AI infrastructure scarcity, one composite index, read against the token index, SAVRN's own cost basis and the frontier ceiling.
Open the scarcity tracker →
Have a specific address? Site Intelligence reads all seven references against one location, county, utility and grid operator included, and cites every finding.
Check an address →