1 live offerings · 1 tracked regions
Publicly listed 1
Loading
AcceleratorsGPU0018
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
GDDR is graphics memory mounted on the board around the GPU rather than stacked on the package. It is widely used on graphics, rendering and cost-efficient inference cards.
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 24 billion parameters, before activations and cache. This is arithmetic on the recorded VRAM, not a benchmark.
Ampere brought third-generation Tensor Cores, TF32 precision and Multi-Instance GPU (MIG) to its data-center parts. (General context about the architecture, not a statement about this specific card.)
Launched in 2020. 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.
PCIe cards slot into standard server and workstation expansion slots, which suits a wide range of systems.
TDP is the rated thermal design power: the planning figure for power delivery and cooling in a server.
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 03 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 $1.09 USD / GPU-hour
Providers represented by current Kovara-tracked offerings.
1 live offerings · 1 tracked regions
Publicly listed 1
1 live offerings · 1 tracked regions
Self-serve 1
4 live offerings · 1 tracked regions
Unknown 4
1 live offerings · 1 tracked regions
Published starting price; capacity unconfirmed 1
2 live offerings · 1 tracked regions
Publicly listed 1 · Self-serve 1
1 live offerings · 1 tracked regions
Exact current public floor not reverified 1
Regions represented by current offerings; this does not describe complete global availability.
4 live offerings · 1 providers
Region profile →6 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-0196Runpod | Unspecified, Multiple / Unspecified | NVIDIA A40 48GB1 GPU configuration · 1 GPU per node | Publicly listedConfirmed capacity: Not confirmed | $0.49 USD / GPU-hourOriginal: USD 0.49 GPU/hour | On-DemandPod GPU instance · Per second / hourly equivalent | Confidence AChecked 20 days agoConfirmed 29 Sept 2026Source: SRC000022 | Inspect |
| OFR-0303Denvr Data | Unspecified, Multiple / Unspecified | NVIDIA A40 48GB4 GPU configuration · 4 GPUs per node | Publicly listedConfirmed capacity: Not confirmed | $0.65 USD / GPU-hourOriginal: USD 0.65 GPU/hour | On-DemandGPU platform · Per-minute billing | Confidence AChecked 25 days agoConfirmed 23 Sept 2026Source: SRC000038 | Inspect |
| OFR-0197Runpod | Unspecified, Multiple / Unspecified | NVIDIA A40 48GB1 GPU configuration · 1 GPU per node | Self-serveConfirmed capacity: Not confirmed | $1.22 USD / GPU-hourOriginal: USD 1.22 GPU/hour | ServerlessServerless GPU worker · Per second / hourly equivalent | Confidence AChecked 20 days agoConfirmed 29 Sept 2026Source: SRC000022 | Inspect |
| OFR-0323E2E Networks | Unspecified, Multiple / Unspecified | NVIDIA A40 48GBConfiguration quantity not specified | Self-serveConfirmed capacity: Not confirmed | $1.44 USD / GPU-hourOriginal: USD 1.44 GPU/hour | On-DemandGPU instance · Per-minute billing | Confidence AChecked 20 days agoConfirmed 29 Sept 2026Source: SRC000040 | Inspect |
| OFR-0251TensorDock | Unspecified, Multiple / Unspecified | NVIDIA A40 48GB1 GPU configuration · 1 GPU per node | Exact current public floor not reverifiedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: Not available | On-DemandMarketplace GPU VM · Dynamic marketplace pricing | Confidence BChecked 20 days agoConfirmed 23 Sept 2026Source: SRC000035 | Inspect |
| OFF000369Exoscale | Frankfurt, Germany | NVIDIA A40 48GBMinimum 1 GPU configuration · 1 GPU per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 1.05 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked 16 days agoConfirmed 02 Oct 2026Source: SRC000168 | Inspect |
| OFR-0362GreenNode | Unspecified, Multiple / Unspecified | NVIDIA A40 48GBConfiguration quantity not specified | Published starting price; capacity unconfirmedConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: USD 860.00 per 30 days | UnknownGPU instance · 30-day | Confidence AChecked 25 days agoConfirmed 24 Sept 2026Source: SRC000046 | Inspect |
| OFF000370Exoscale | Frankfurt, Germany | NVIDIA A40 48GBMinimum 2 GPU configuration · 2 GPUs per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 2.09 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked 16 days agoConfirmed 02 Oct 2026Source: SRC000168 | Inspect |
| OFF000371Exoscale | Frankfurt, Germany | NVIDIA A40 48GBMinimum 4 GPU configuration · 4 GPUs per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 4.18 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked 16 days agoConfirmed 02 Oct 2026Source: SRC000168 | Inspect |
| OFF000372Exoscale | Frankfurt, Germany | NVIDIA A40 48GBMinimum 8 GPU configuration · 8 GPUs per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 8.36 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked 16 days agoConfirmed 02 Oct 2026Source: SRC000168 | Inspect |
Recorded hardware attributes. Not tracked is distinct from No.
Public evidence and confidence signals supporting Kovara-tracked coverage.