H100 SXM rental price: from $2.25/hr
The cheapest on-demand H100 SXM is $2.25 an hour at TensorDock, checked 21 November 2026; the typical provider charges $3.49.
- Spot $1.91 TensorDock, interruptible
- Reserved $2.72 Hyperstack, term not published
- Serverless $3.95 Modal, per second
- In stock $4.39 RunPod, stock reported
- Provider median $3.49
- Highest on-demand $12.29
- 22 offers
- 12 providers
- Cheapest unchanged since 18 July
- Same offers, 90 days: +24.0%
- Source tensordock.com
H100 SXM offers: 22 configurations from 12 providers
More filters: provider, memory, node size, stock, place, CPU and RAM, sort, price unit
“Machines in” filters the 8 of 22 offers whose provider says where the machine is; the others are under “Place not stated”.
“vCPUs per GPU” and “RAM per GPU” use the 6 of 22 offers whose provider publishes the machine's vCPUs or RAM for a stated number of GPUs (AWS, Lambda and RunPod); an offer that does not publish them never matches a minimum. A GiB counts as 1.07 GB.
Hyperstack, Lambda, Massed Compute, Modal and Thunder Compute put GPUs in North America on their own pages, but none of their offers here says so.
Hyperstack, Lambda, Modal and Verda (DataCrunch) put GPUs in Europe on their own pages, but none of their offers here says so.
Jarvislabs, Lambda and Modal put GPUs in Asia and Oceania on their own pages, but none of their offers here says so.
- Rent →
TensorDock1 configuration$2.25on-demandNot reportedOn-demand $2.25Spot & community $1.91
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM spot / interruptible $1.91 node size not published checked 21 November 2026 · source H100 SXM on-demand $2.25 node size not published checked 21 November 2026 · source TensorDock H100 SXM pricing · TensorDock: every GPU, fees and billing
- Rent →
Massed Compute1 configuration$3.14on-demandNot reportedOn-demand $3.14
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM on-demand $3.14 single GPU checked 21 November 2026 · source Massed Compute H100 SXM pricing · Massed Compute: every GPU, fees and billing
- Rent →
Hyperstack1 configuration$3.20on-demandNot reportedOn-demand $3.20Reserved $2.72
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM reserved / committedTerm and minimum volume not published in this table; confirm reservation quote $2.72 node size not published checked 21 November 2026 · source H100 SXM on-demand $3.20 node size not published checked 21 November 2026 · source Hyperstack H100 SXM pricing · Hyperstack: every GPU, fees and billing
- Rent →
Thunder Compute1 configuration$3.20on-demandNot reportedOn-demand $3.20
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM on-demand $3.20 single GPU checked 21 November 2026 · source Thunder Compute H100 SXM pricing · Thunder Compute: every GPU, fees and billing
- Rent →
Scaleway1 configuration$3.48on-demandNot reportedOn-demand $3.48Paris, France
Configuration Billing $/GPU-hour Node Stock · checked H100 SXMParis, France on-demand $3.48 8 GPUs · $27.84/hr per node checked 21 November 2026 · source Scaleway H100 SXM pricing · Scaleway: every GPU, fees and billing
- Rent →
Jarvislabs1 configuration$3.49on-demandNot reportedOn-demand $3.49
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM on-demand $3.49 node size not published checked 21 November 2026 · source Jarvislabs H100 SXM pricing · Jarvislabs: every GPU, fees and billing
- Rent →
RunPod1 configuration$4.39secure cloudIn stock: highOn-demand $4.39Spot & community $2.69
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM community cloud $2.69 node size not published checked 21 November 2026 · source H100 SXM secure cloud $4.39 node size not published In stock: high · checked 21 November 2026 · source RunPod H100 SXM pricing · RunPod: every GPU, fees and billing
- Rent →
Lambda3 configurations$4.69on-demandNot reportedOn-demand $4.69
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM4x NVIDIA H100 SXM on-demand $4.69 4 GPUs · $18.76/hr per node checked 21 November 2026 · source H100 SXM2x NVIDIA H100 SXM on-demand $4.79 2 GPUs · $9.58/hr per node checked 21 November 2026 · source H100 SXM1x NVIDIA H100 SXM on-demand $4.89 single GPU checked 21 November 2026 · source Lambda H100 SXM pricing · Lambda: every GPU, fees and billing
- Rent →
Verda (DataCrunch)1 configuration$5.51on-demandNot reportedOn-demand $5.51Spot & community $2.75
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM spot / interruptible $2.75 single GPU checked 21 November 2026 · source H100 SXM on-demand $5.51 single GPU checked 21 November 2026 · source Verda (DataCrunch) H100 SXM pricing · Verda (DataCrunch): every GPU, fees and billing
- Rent →
AWS1 configuration$6.88on-demandNot reportedOn-demand $6.88Spot & community $7.22US East (N. Virginia)
- Rent →
Microsoft Azure2 configurations$11.06on-demandNot reportedOn-demand $11.06Spot & community $2.04East US
Configuration Billing $/GPU-hour Node Stock · checked H100 SXMStandard_ND96is_H100_v5East US spot / interruptible $2.04 8 GPUs · $16.35/hr per node checked 21 November 2026 · source H100 SXMStandard_ND96isr_H100_v5East US spot / interruptible $2.27 8 GPUs · $18.17/hr per node checked 21 November 2026 · source H100 SXMStandard_ND96is_H100_v5East US on-demand $11.06 8 GPUs · $88.49/hr per node checked 21 November 2026 · source H100 SXMStandard_ND96isr_H100_v5East US on-demand $12.29 8 GPUs · $98.32/hr per node checked 21 November 2026 · source Microsoft Azure H100 SXM pricing · Microsoft Azure: every GPU, fees and billing
- Rent →
Modal1 configuration$3.95serverlessNot reportedServerless $3.95
Configuration Billing $/GPU-hour Node Stock · checked H100 SXM serverless $3.95 node size not published checked 21 November 2026 · source
11 providers with on-demand offers, fresh prices first, then cheapest. Tap a provider for its configurations.
| GPU / provider | Billing | $/GPU-hour | Node · total per hour | Rent |
|---|---|---|---|---|
| H100 SXM · TensorDock80 GB | on-demand | $2.25 | – | Rent → |
| H100 SXM · Massed Compute80 GB | on-demand | $3.14 | 1× $3.14 | Rent → |
| H100 SXM · Hyperstack80 GB | on-demand | $3.20 | – | Rent →checked 21 NovDetails
|
| H100 SXM · Thunder Compute80 GB | on-demand | $3.20 | 1× $3.20 | Rent → |
| H100 SXM · Scaleway80 GB | on-demand | $3.48 | 8× $27.84 | Rent → |
| H100 SXM · Jarvislabs80 GB | on-demand | $3.49 | – | Rent → |
| H100 SXM · RunPod80 GB | secure cloud | $4.39+10.0% in 30 days | – | In stock: highRent →checked 21 NovDetails
|
| H100 SXM · Lambda80 GB · 4x NVIDIA H100 SXM | on-demand | $4.69+14.7% in 30 days | 4× $18.76 | Rent →checked 21 NovDetails
|
| H100 SXM · Lambda80 GB · 2x NVIDIA H100 SXM | on-demand | $4.79+14.3% in 30 days | 2× $9.58 | Rent →checked 21 NovDetails
|
| H100 SXM · Lambda80 GB · 1x NVIDIA H100 SXM | on-demand | $4.89+14.0% in 30 days | 1× $4.89 | Rent →checked 21 NovDetails
|
| H100 SXM · Verda (DataCrunch)80 GB | on-demand | $5.51+23.2% in 30 days | 1× $5.51 | Rent → |
| H100 SXM · AWS80 GB · p5.48xlarge | on-demand | $6.88 | 8× $55.04 | Rent →checked 21 NovDetails
|
| H100 SXM · Microsoft Azure80 GB · Standard_ND96is_H100_v5 | on-demand | $11.06 | 8× $88.49 | Rent →checked 21 NovDetails
|
| H100 SXM · Microsoft Azure80 GB · Standard_ND96isr_H100_v5 | on-demand | $12.29 | 8× $98.32 | Rent →checked 21 NovDetails
|
| H100 SXM · TensorDock80 GB | spot / interruptible | $1.91 | – | Rent → |
| H100 SXM · Microsoft Azure80 GB · Standard_ND96is_H100_v5 | spot / interruptible | $2.04 | 8× $16.35 | Rent →checked 21 NovDetails
|
| H100 SXM · Microsoft Azure80 GB · Standard_ND96isr_H100_v5 | spot / interruptible | $2.27 | 8× $18.17 | Rent →checked 21 NovDetails
|
| H100 SXM · RunPod80 GB | community cloud | $2.69 | – | Rent → |
| H100 SXM · Hyperstack80 GB | reserved / committedTerm and minimum volume not published in this table; confirm reservation quoteStarting rate | $2.72 | – | Rent → |
| H100 SXM · Verda (DataCrunch)80 GB | spot / interruptible | $2.75 | 1× $2.75 | Rent → |
| H100 SXM · Modal80 GB | serverless | $3.95$0.001097/sec | – | Rent →checked 21 NovDetails
|
| H100 SXM · AWS80 GB · p5.48xlarge | spot / interruptible | $7.22 | 8× $57.76 | Rent →checked 21 NovDetails
|
Try another provider or reduce the minimum memory.
A spot machine can be stopped whenever the provider needs it back: how much notice each provider gives, and what happens to the disk.
Fresh prices come first. Within each billing type, prices run lowest first, with unavailable offers last. Commercial relationships never affect ordering. How we collect and rank prices.
Same GPU, provider, billing and price on several hosts or listings: one row, with the count. – means the provider does not state it.
Monthly view assumes 730 hours of continuous use. Daily view assumes 24 hours.
Cite this data
GPU Rental Prices, "H100 SXM rental price: from $2.25/GPU-hour", prices checked 21 November 2026, https://gpurentalprices.com/h100-sxm
Quote it with a link to this page. Downloads of the whole dataset are on Data & API.
On-demand and secure-cloud rates are capacity the provider runs, with no reservation. Spot and community capacity can be interrupted. A marketplace host is one independent owner's server listed on a marketplace. Prices are per GPU per hour, before storage and data transfer. Every row shows the date it was checked.
H100 SXM price history and range
- Cheapest on-demand$2.25(unchanged since 18 July)
- Provider median$3.49
- Cheapest spot$1.91
H100 SXM price range by billing type
Per GPU-hour, latest checked offers. Axis $0 to $20.
CheapestProvider median (each provider's cheapest rate counted once; shown with three or more providers)
Price changes
- 30 days: +5.6% across 13 matching on-demand configurations since 22 October 2026.
- 90 days: +24.0% across 5 matching on-demand configurations since 23 August 2026.
Changes use the geometric mean of price ratios for the same provider, variant, tier, GPU count, instance and region (RunPod's place is where that day's stock sat, so it is not matched). New listings are not counted as price changes. Full ledger on the history hub.
H100 SXM daily figures (140 days)
| Date | Cheapest on-demand | Provider median | Cheapest spot |
|---|---|---|---|
| 2026-11-21 | $2.25 | $3.49 | $1.91 |
| 2026-11-20 | $2.25 | $3.49 | $1.91 |
| 2026-11-19 | $2.25 | $3.49 | $1.91 |
| 2026-11-18 | $2.25 | $3.49 | $1.91 |
| 2026-11-17 | $2.25 | $3.49 | $1.91 |
| 2026-11-16 | $2.25 | $3.50 | $1.91 |
| 2026-11-15 | $2.25 | $3.50 | $1.91 |
| 2026-11-14 | $2.25 | $3.50 | $1.91 |
| 2026-11-13 | $2.25 | $3.51 | $1.91 |
| 2026-11-12 | $2.25 | $3.52 | $1.91 |
| 2026-11-11 | $2.25 | $3.55 | $1.91 |
| 2026-11-10 | $2.25 | $3.54 | $1.91 |
| 2026-11-09 | $2.25 | $3.54 | $1.91 |
| 2026-11-08 | $2.25 | $3.54 | $1.91 |
| 2026-11-07 | $2.25 | $3.54 | $1.91 |
| 2026-11-06 | $2.25 | $3.55 | $1.91 |
| 2026-11-05 | $2.25 | $3.57 | $1.91 |
| 2026-11-04 | $2.25 | $3.54 | $1.91 |
| 2026-11-03 | $2.25 | $3.74 | $1.91 |
| 2026-11-02 | $2.25 | $3.74 | $1.91 |
| 2026-11-01 | $2.25 | $3.74 | $1.91 |
| 2026-10-31 | $2.25 | $3.74 | $1.91 |
| 2026-10-30 | $2.25 | $3.99 | $1.91 |
| 2026-10-29 | $2.25 | $3.77 | $1.91 |
| 2026-10-28 | $2.25 | $3.77 | $1.91 |
| 2026-10-27 | $2.25 | $3.78 | $1.91 |
| 2026-10-26 | $2.25 | $3.78 | $1.91 |
| 2026-10-25 | $2.25 | $3.57 | $1.91 |
| 2026-10-24 | $2.25 | $3.57 | $1.91 |
| 2026-10-23 | $2.25 | $3.58 | $1.91 |
| 2026-10-22 | $2.25 | $3.59 | $1.91 |
| 2026-10-21 | $2.25 | $3.58 | $1.91 |
| 2026-10-20 | $2.25 | $3.58 | $1.91 |
| 2026-10-19 | $2.25 | $3.59 | $1.91 |
| 2026-10-18 | $2.25 | $3.59 | $1.91 |
| 2026-10-17 | $2.25 | $3.59 | $1.91 |
| 2026-10-16 | $2.25 | $3.58 | $1.91 |
| 2026-10-15 | $2.25 | $3.57 | $1.91 |
| 2026-10-14 | $2.25 | $3.57 | $1.91 |
| 2026-10-13 | $2.25 | $3.57 | $1.91 |
| 2026-10-12 | $2.25 | $3.58 | $1.91 |
| 2026-10-11 | $2.25 | $3.58 | $1.91 |
| 2026-10-10 | $2.25 | $3.58 | $1.91 |
| 2026-10-09 | $2.25 | $3.57 | $1.91 |
| 2026-10-08 | $2.25 | $3.49 | $1.91 |
| 2026-10-07 | $2.25 | $3.49 | $1.91 |
| 2026-10-06 | $2.25 | $3.49 | $1.91 |
| 2026-10-05 | $2.25 | $3.49 | $1.91 |
| 2026-10-04 | $2.25 | $3.49 | $1.91 |
| 2026-10-03 | $2.25 | $3.49 | $1.91 |
| 2026-10-02 | $2.25 | $3.49 | $1.91 |
| 2026-10-01 | $2.25 | $3.49 | $1.91 |
| 2026-09-30 | $2.25 | $3.54 | $1.91 |
| 2026-09-29 | $2.25 | $3.52 | $1.91 |
| 2026-09-28 | $2.25 | $3.52 | $1.91 |
| 2026-09-27 | $2.25 | $3.49 | $1.91 |
| 2026-09-26 | $2.25 | $3.48 | $1.91 |
| 2026-09-25 | $2.25 | $3.45 | $1.91 |
| 2026-09-24 | $2.25 | $3.42 | $1.91 |
| 2026-09-23 | $2.25 | $3.38 | $1.91 |
| 2026-09-22 | $2.25 | $3.37 | $1.91 |
| 2026-09-21 | $2.25 | $3.35 | $1.91 |
| 2026-09-20 | $2.25 | $3.35 | $1.91 |
| 2026-09-19 | $2.25 | $3.35 | $1.91 |
| 2026-09-18 | $2.25 | $3.30 | $1.91 |
| 2026-09-17 | $2.25 | $3.28 | $1.91 |
| 2026-09-16 | $2.25 | $3.28 | $1.91 |
| 2026-09-15 | $2.25 | $3.27 | $1.91 |
| 2026-09-14 | $2.25 | $3.25 | $1.91 |
| 2026-09-13 | $2.25 | $3.25 | $1.91 |
| 2026-09-12 | $2.25 | $3.25 | $1.91 |
| 2026-09-11 | $2.25 | $3.25 | $1.91 |
| 2026-09-10 | $2.25 | $3.25 | $1.91 |
| 2026-09-09 | $2.25 | $3.25 | $1.91 |
| 2026-09-08 | $2.25 | $3.23 | $1.91 |
| 2026-09-07 | $2.25 | $3.23 | $1.91 |
| 2026-09-06 | $2.25 | $3.23 | $1.91 |
| 2026-09-05 | $2.25 | $3.23 | $1.91 |
| 2026-09-04 | $2.25 | $3.23 | $1.91 |
| 2026-09-03 | $2.25 | $3.23 | $1.91 |
| 2026-09-02 | $2.25 | $3.23 | $1.91 |
| 2026-09-01 | $2.25 | $3.25 | $1.91 |
| 2026-08-31 | $2.25 | $3.25 | $1.91 |
| 2026-08-30 | $2.25 | $3.25 | $1.91 |
| 2026-08-29 | $2.25 | $3.25 | $1.91 |
| 2026-08-28 | $2.25 | $3.25 | $1.91 |
| 2026-08-27 | $2.25 | $3.25 | $1.91 |
| 2026-08-26 | $2.25 | $3.25 | $1.91 |
| 2026-08-25 | $2.25 | $3.25 | $1.91 |
| 2026-08-24 | $2.25 | $3.25 | $1.91 |
| 2026-08-23 | $2.25 | $3.25 | $1.91 |
| 2026-08-22 | $2.25 | $3.25 | $1.91 |
| 2026-08-21 | $2.25 | $3.25 | $1.91 |
| 2026-08-20 | $2.25 | $3.25 | $1.91 |
| 2026-08-19 | $2.25 | $3.25 | $1.91 |
| 2026-08-18 | $2.25 | $3.25 | $1.91 |
| 2026-08-17 | $2.25 | $3.25 | $1.91 |
| 2026-08-16 | $2.25 | $3.25 | $1.91 |
| 2026-08-15 | $2.25 | $3.25 | $1.91 |
| 2026-08-14 | $2.25 | $3.25 | $1.91 |
| 2026-08-13 | $2.25 | $3.25 | $1.91 |
| 2026-08-12 | $2.25 | $3.25 | $1.91 |
| 2026-08-11 | $2.25 | $3.25 | $1.91 |
| 2026-08-10 | $2.25 | $3.25 | $1.91 |
| 2026-08-09 | $2.25 | $3.25 | $1.91 |
| 2026-08-08 | $2.25 | $3.25 | $1.91 |
| 2026-08-07 | $2.25 | $3.25 | $1.91 |
| 2026-08-06 | $2.25 | $3.25 | $1.91 |
| 2026-08-05 | $2.25 | $3.25 | $1.91 |
| 2026-08-04 | $2.25 | $3.20 | $1.91 |
| 2026-08-03 | $2.25 | $3.20 | $1.91 |
| 2026-08-02 | $2.25 | $3.20 | $1.91 |
| 2026-08-01 | $2.25 | $3.20 | $1.91 |
| 2026-07-31 | $2.25 | $3.20 | $1.91 |
| 2026-07-30 | $2.25 | $3.20 | $1.91 |
| 2026-07-29 | $2.25 | $3.20 | $1.91 |
| 2026-07-28 | $2.25 | $3.20 | $1.91 |
| 2026-07-27 | $2.25 | $3.20 | $1.91 |
| 2026-07-26 | $2.25 | $3.20 | $1.91 |
| 2026-07-25 | $2.25 | $3.20 | $1.91 |
| 2026-07-24 | $2.25 | $3.20 | $1.91 |
| 2026-07-23 | $2.25 | $3.20 | $1.91 |
| 2026-07-22 | $2.25 | $3.20 | $1.91 |
| 2026-07-21 | $2.25 | $3.20 | $1.91 |
| 2026-07-20 | $2.25 | $3.20 | $1.91 |
| 2026-07-19 | $2.25 | $3.17 | $1.91 |
| 2026-07-18 | $2.25 | $3.17 | $1.91 |
| 2026-07-17 | $2.69 | $3.23 | $2.04 |
| 2026-07-16 | $2.69 | $3.23 | $2.04 |
| 2026-07-15 | $2.69 | $3.23 | $2.04 |
| 2026-07-14 | $2.69 | $3.23 | $2.04 |
| 2026-07-13 | $2.69 | $3.23 | $2.04 |
| 2026-07-12 | $2.69 | $3.23 | $2.04 |
| 2026-07-11 | $2.69 | $3.23 | $2.04 |
| 2026-07-10 | $2.69 | $3.23 | $2.04 |
| 2026-07-09 | $2.69 | $3.23 | $2.04 |
| 2026-07-08 | $2.69 | $3.23 | $2.04 |
| 2026-07-07 | $2.69 | $3.23 | $2.04 |
| 2026-07-06 | $2.69 | $3.23 | – |
| 2026-07-05 | $3.29 | – | $2.69 |
H100 SXM prices by provider and billing type
Each provider's cheapest rate in each billing type, per GPU-hour. The provider name opens its H100 SXM page with node sizes, fees and its own history.
| Provider | On-demand | Spot & community | Reserved | Serverless |
|---|---|---|---|---|
| TensorDock | $2.25 | $1.91 | – | – |
| Massed Compute | $3.14 | – | – | – |
| Hyperstack | $3.20 | – | $2.72 | – |
| Thunder Compute | $3.20 | – | – | – |
| Scaleway | $3.48 | – | – | – |
| Jarvislabs | $3.49 | – | – | – |
| RunPod | $4.39 | $2.69 | – | – |
| Lambda | $4.69 | – | – | – |
| Verda (DataCrunch) | $5.51 | $2.75 | – | – |
| AWS | $6.88 | $7.22 | – | – |
| Microsoft Azure | $11.06 | $2.04 | – | – |
| Modal | – | – | – | $3.95 |
H100 SXM hardware specs
| Architecture | Hopper |
|---|---|
| VRAM | 80 GB |
| FP16 TFLOPS (dense) | 989 |
| Memory bandwidth | 3,350 GB/s |
| TDP | 700 W |
| Interconnect | NVLink 900GB/s |
| Launched | 2022 |
Manufacturer datasheet figures for the H100 SXM. Source: resources.nvidia.com.
H100 SXM variants
LLM memory estimates for this GPU
These models have an estimated inference footprint within 80 GB at one or more precisions. The scenario uses 4,096 tokens, batch 1, a 16-bit KV cache and 20% runtime headroom. A memory fit does not establish speed or software compatibility.
| Model | 16-bit GB | 8-bit GB | 4-bit GB |
|---|---|---|---|
| GLM 4.5 Air | 266.1 | 133.5 | 67.2 ✓ |
| Qwen2.5 72B | 176.1 | 88.9 | 45.2 ✓ |
| DeepSeek R1 Distill Llama 70B | 170.9 | 86.3 | 43.9 ✓ |
| Mixtral 8x7B v0.1 | 112.7 | 56.7 ✓ | 28.7 ✓ |
| Qwen2.5 32B | 79.9 ✓ | 40.6 ✓ | 20.9 ✓ |
| Qwen2.5 Coder 32B | 79.9 ✓ | 40.6 ✓ | 20.9 ✓ |
| DeepSeek R1 Distill Qwen 32B | 79.9 ✓ | 40.6 ✓ | 20.9 ✓ |
| QwQ 32B | 79.9 ✓ | 40.6 ✓ | 20.9 ✓ |
| Qwen3 32B | 79.9 ✓ | 40.6 ✓ | 20.9 ✓ |
| Qwen3 30B A3B | 73.8 ✓ | 37.1 ✓ | 18.8 ✓ |
| Qwen3 30B A3B 2507 | 73.8 ✓ | 37.1 ✓ | 18.8 ✓ |
| Qwen3 Coder 30B A3B | 73.8 ✓ | 37.1 ✓ | 18.8 ✓ |
| Mistral Small 3.2 24B 2506 | 58.4 ✓ | 29.6 ✓ | 15.2 ✓ |
| Mistral Small 24B 2501 | 57.4 ✓ | 29.1 ✓ | 14.9 ✓ |
| Devstral Small 2507 | 57.4 ✓ | 29.1 ✓ | 14.9 ✓ |
| Qwen2.5 14B | 36.4 ✓ | 18.7 ✓ | 9.8 ✓ |
| Qwen2.5 Coder 14B | 36.4 ✓ | 18.7 ✓ | 9.8 ✓ |
| DeepSeek R1 Distill Qwen 14B | 36.4 ✓ | 18.7 ✓ | 9.8 ✓ |
| Qwen3 14B | 36.2 ✓ | 18.5 ✓ | 9.7 ✓ |
| phi 4 | 36.2 ✓ | 18.6 ✓ | 9.8 ✓ |
| Phi 4 reasoning | 36.2 ✓ | 18.6 ✓ | 9.8 ✓ |
| Phi 4 reasoning plus | 36.2 ✓ | 18.6 ✓ | 9.8 ✓ |
| Mistral Nemo 2407 | 30.2 ✓ | 15.5 ✓ | 8.2 ✓ |
| Qwen3 8B | 20.4 ✓ | 10.6 ✓ | 5.6 ✓ |
| DeepSeek R1 Distill Llama 8B | 19.9 ✓ | 10.3 ✓ | 5.5 ✓ |
| Qwen2.5 7B | 18.6 ✓ | 9.4 ✓ | 4.9 ✓ |
| Qwen2.5 Coder 7B | 18.6 ✓ | 9.4 ✓ | 4.9 ✓ |
| DeepSeek R1 Distill Qwen 7B | 18.6 ✓ | 9.4 ✓ | 4.9 ✓ |
| Mistral 7B v0.3 | 18.0 ✓ | 9.3 ✓ | 5.0 ✓ |
| Qwen3 4B | 10.4 ✓ | 5.6 ✓ | 3.1 ✓ |
| Qwen3 4B 2507 | 10.4 ✓ | 5.6 ✓ | 3.1 ✓ |
| Qwen2.5 3B | 7.6 ✓ | 3.9 ✓ | 2.0 ✓ |
| Qwen2.5 Coder 3B | 7.6 ✓ | 3.9 ✓ | 2.0 ✓ |
| Qwen3 1.7B | 5.4 ✓ | 3.0 ✓ | 1.8 ✓ |
| DeepSeek R1 Distill Qwen 1.5B | 4.4 ✓ | 2.3 ✓ | 1.2 ✓ |
| Qwen2.5 1.5B | 3.8 ✓ | 2.0 ✓ | 1.1 ✓ |
| Qwen2.5 Coder 1.5B | 3.8 ✓ | 2.0 ✓ | 1.1 ✓ |
| Qwen3 0.6B | 2.4 ✓ | 1.5 ✓ | 1.0 ✓ |
| Qwen2.5 0.5B | 1.2 ✓ | 0.7 ✓ | 0.4 ✓ |
| Qwen2.5 Coder 0.5B | 1.2 ✓ | 0.7 ✓ | 0.4 ✓ |
Models that need more than one H100 SXM
- Qwen3 235B A22B at 16-bit needs about 565 GB with the same context and headroom, so at least 8 cards. The cheapest H100 SXM node with a published size that holds it: 8× at $27.84/hr (Scaleway, $3.48 per GPU-hour). At 8 or 4 bit it needs less; the model page recalculates.
- Qwen3 235B A22B 2507 at 16-bit needs about 565 GB with the same context and headroom, so at least 8 cards. The cheapest H100 SXM node with a published size that holds it: 8× at $27.84/hr (Scaleway, $3.48 per GPU-hour). At 8 or 4 bit it needs less; the model page recalculates.
- Mixtral 8x22B v0.1 at 16-bit needs about 339 GB with the same context and headroom, so at least 5 cards. The cheapest H100 SXM node with a published size that holds it: 8× at $27.84/hr (Scaleway, $3.48 per GPU-hour). At 8 or 4 bit it needs less; the model page recalculates.
Compare model memory requirements or change context and memory assumptions on a model's page.
When to choose the H100 SXM, and what to pick instead
Start from memory: your model, its context and your batch must fit in 80 GB (the model-fit table above estimates it). Then the H100 SXM is the right card unless one of these fits your job better.
- Pick the A100 (80 GB, from $1.09/hr) when your model fits in 80 GB and you can wait longer: it costs 52% less per hour.
- Pick the H200 (141 GB, from $3.20/hr) when the model or its cache does not fit in 80 GB.
- Pick the B200 (180 GB, from $5.00/hr) when finishing sooner is worth paying more: it has 2.3 times the H100 SXM's 16-bit compute.
Rent or buy the H100 SXM
Confirmed H100 SXM purchase listings start at $29,900 (H100 NVL 94GB, NetworkOutlet, checked 21 November 2026). That sum rents about 13,289 GPU-hours at the cheapest on-demand rate of $2.25 (checked 21 November 2026), about 6.4 years at 40 hours a week, before electricity, hosting and resale.
Compare H100 SXM purchase prices and quote availability, then enter your usage and electricity costs in the rent-versus-buy calculator. Purchase listings have their own variants, stock status and update dates. Compare two GPUs.
H100 SXM questions
- How much does it cost to rent an H100 SXM?
- The cheapest H100 SXM rental we have is $2.25 per GPU-hour on TensorDock (on-demand), checked 21 November 2026. The median of each provider's cheapest on-demand rate is $3.49 across 11 providers.
- How much VRAM does an H100 SXM have?
- 80 GB.
- Where is the H100 SXM cheapest right now?
- TensorDock has the cheapest checked on-demand rate on 21 November 2026 at $2.25 per GPU-hour; interruptible spot or community capacity starts at $1.91 (TensorDock).
GPUs near the H100 SXM
- GH20096 GB$2.59/hrcheapest on-demand
- A10040-80 GB$0.89/hrcheapest on-demand
- H200141-143 GB$3.20/hrcheapest on-demand
- MI300X192 GB$2.59/hrcheapest on-demand
- RTX PRO 600096 GB$0.80/hrcheapest on-demand
Related
All 80 GB+ GPUs · GPUs under $2/hr · Price movers this week · TensorDock review