8 live offerings · 1 tracked regions
Published configuration; live capacity unconfirmed 8
Loading
Loading workspaces…
engineering · GPU0092
Specifications with their source, supported precisions, interconnect, benchmark coverage and which published models fit NVIDIA H200 141GB — provider does not specify SXM/NVL.
| Specification | Value | Source | Kovara registry |
|---|---|---|---|
| Memory capacity | 141GB | Kovara registry · verified 2026-09-25 | used |
| Memory type | HBM3E | Kovara registry · verified 2026-09-25 | used |
| Memory bandwidth | 4.8TB/s | Kovara registry · verified 2026-09-25 | used |
| Board power (TDP) | — | Not published | — |
| Form factor | — | Not published | — |
| NVLink | Yes | Kovara registry · verified 2026-09-25 | used |
| MIG partitioning | Yes | Kovara registry · verified 2026-09-25 | used |
No public benchmark runs yet
Nobody has published a hardware run on this accelerator. Measured numbers appear here once a run of a compute repository is published.
AcceleratorsGPU0092
NVIDIA
Core hardware and lifecycle, drawn from Kovara master data.
What the recorded specifications mean in practice, and what this accelerator is used for.
Specs, explained
HBM (High Bandwidth Memory) is stacked in layers and sits right next to the processor on the same package. That short, wide path is what gives accelerators with HBM their memory bandwidth, so they are usually chosen for work limited by how fast data reaches the cores, such as large-model inference and HPC simulation.
VRAM is the working space on the card. Model weights, activations and datasets have to fit in it, or be split across several GPUs. Rule of thumb for the weights alone at 16-bit precision (about 2 GB per billion parameters): up to roughly 70 billion parameters, before activations and cache. This is arithmetic on the recorded VRAM, not a benchmark.
Hopper added the Transformer Engine, which mixes FP8 and 16-bit math for transformer models, and fourth-generation Tensor Cores. (General context about the architecture, not a statement about this specific card.)
Launched in 2023. Kovara's records list it as a previous-generation part.
NVLink is a direct GPU-to-GPU link. Where it is supported, GPUs in the same server can exchange data faster than over PCIe alone, which matters when one training job is spread across several GPUs.
Multi-Instance GPU (MIG) can divide one physical GPU into isolated slices, each with its own memory and compute, so several smaller workloads can share a card.
These are the uses Kovara has recorded for this part. The notes explain the workload in general terms; they are not performance claims.
Figures are Kovara master data, last verified 25 Sept 2026. Explanations describe the technology in general; they are not benchmarks and say nothing about price or availability.
Explore this accelerator on ComputePublic Kovara observations; not a claim of complete or universal market coverage. Incompatible raw price units are excluded.
Observed GPU-hour range
USD / GPU-hour
Spread $3.00 USD / GPU-hour
Providers represented by current Kovara-tracked offerings.
8 live offerings · 1 tracked regions
Published configuration; live capacity unconfirmed 8
1 live offerings · 1 tracked regions
Self-serve 1
1 live offerings · 1 tracked regions
Published starting price; capacity unconfirmed 1
1 live offerings · 1 tracked regions
Publicly listed 1
1 live offerings · 1 tracked regions
Self-serve 1
1 live offerings · 1 tracked regions
Quote or tariff confirmation required; live capacity unconfirmed 1
1 live offerings · 1 tracked regions
Live marketplace — query required 1
Regions represented by current offerings; this does not describe complete global availability.
8 live offerings · 1 providers
Region profile →1 live offerings · 1 providers
Region profile →5 live offerings · 5 providers
Current configurations, recorded availability, confirmed capacity and the evidence behind each price.
| Provider | Region | Configuration | Availability | Price | Commercial model | Trust / evidence | Action |
|---|---|---|---|---|---|---|---|
| OFR-0327E2E Networks | Unspecified, Multiple / Unspecified | NVIDIA H200 141GB — provider does not specify SXM/NVLConfiguration quantity not specified | Self-serveConfirmed capacity: Not confirmed | $5.29 USD / GPU-hourOriginal: USD 5.29 GPU/hour | On-DemandGPU instance · Per-minute billing | Confidence AChecked 16 days agoConfirmed 29 Sept 2026Source: SRC000040 | Inspect |
| OFR-0216Runpod | Unspecified, Multiple / Unspecified | NVIDIA H200 141GB — provider does not specify SXM/NVL1 GPU configuration · 1 GPU per node | Self-serveConfirmed capacity: Not confirmed | $6.99 USD / GPU-hourOriginal: USD 6.99 GPU/hour | ServerlessServerless GPU worker · Per second / hourly equivalent | Confidence AChecked todayConfirmed 29 Sept 2026Source: SRC000022 | Inspect |
| OFR-0475OVHcloud | Unspecified, Multiple / Unspecified | NVIDIA H200 141GB — provider does not specify SXM/NVL4 GPU configuration · 4 GPUs per node | Publicly listedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 23.12 instance-hour | On-DemandManaged AI GPU resource — product scope unconfirmed · Hourly | Confidence AChecked 16 days agoConfirmed 24 Sept 2026Source: SRC-OVHCLOUD-CATALOG-API-20261031 | Inspect |
| OFR-0693Vast.ai | Unspecified, Multiple / Unspecified | NVIDIA H200 141GB — provider does not specify SXM/NVL1 GPU configuration · 1 GPU per node | Live marketplace — query requiredConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: Not available | On-DemandMarketplace GPU instance · Per-second / dynamic | Confidence BChecked 55 days ago · StaleNot confirmedSource: SRC000052 | Inspect |
| OFR-0875Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL1 GPU per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 119000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0876Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL2 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 239000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0877Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL3 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 359000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0878Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL4 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 479000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0879Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL5 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 599000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0880Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL6 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 719000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0881Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL7 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 839000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0882Biznet Gio | Indonesia (location unspecified), Indonesia | NVIDIA H200 141GB — provider does not specify SXM/NVL8 GPUs per node | Published configuration; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: IDR 959000.00 instance-hour | On-DemandGPU cloud instance · Hourly | Confidence BChecked todayNot confirmedSource: SRC-SEA-20260928-BIZNET-PRICE | Inspect |
| OFR-0907SIAM.AI Cloud | Thailand (location unspecified), Thailand | NVIDIA H200 141GB — provider does not specify SXM/NVLConfiguration quantity not specified | Quote or tariff confirmation required; live capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: Not available | On-DemandGPU cloud service · Quote on request | Confidence BChecked 50 days ago · StaleNot confirmedSource: SRC-SEA-20260928-SIAM-HSERIES | Inspect |
| OFR-0366GreenNode | Unspecified, Multiple / Unspecified | NVIDIA H200 141GB — provider does not specify SXM/NVLConfiguration quantity not specified | Published starting price; capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: USD 3076.00 per 30 days | UnknownGPU instance · 30-day | Confidence AChecked 54 days ago · StaleConfirmed 24 Sept 2026Source: SRC000046 | Inspect |
Recorded hardware attributes. Not tracked is distinct from No.
Public evidence and confidence signals supporting Kovara-tracked coverage.