3 live offerings · 1 tracked regions
Unknown 3
Loading
Loading workspaces…
engineering · GPU0095
Specifications with their source, supported precisions, interconnect, benchmark coverage and which published models fit NVIDIA GeForce RTX 3080 Ti Reference specification.
| Specification | Value | Source | Kovara registry |
|---|---|---|---|
| Memory capacity | 12GB | Kovara registry · verified 2026-09-23 | used |
| Memory type | GDDR6X | Kovara registry · verified 2026-09-23 | used |
| Memory bandwidth | — | Not published | — |
| Board power (TDP) | 350W | Kovara registry · verified 2026-09-23 | used |
| Form factor | PCIe consumer | Kovara registry · verified 2026-09-23 | used |
| NVLink | — | Not published | — |
| MIG partitioning | — | Not published | — |
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.
AcceleratorsGPU0095
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 6 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 2021.
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.
Figures are Kovara master data, last verified 23 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 $0.12 USD / GPU-hour
Providers represented by current Kovara-tracked offerings.
3 live offerings · 1 tracked regions
Unknown 3
4 live offerings · 1 tracked regions
Unconfirmed 4
Regions represented by current offerings; this does not describe complete global availability.
4 live offerings · 1 providers
3 live offerings · 1 providers
Region profile →Current configurations, recorded availability, confirmed capacity and the evidence behind each price.
| Provider | Region | Configuration | Availability | Price | Commercial model | Trust / evidence | Action |
|---|---|---|---|---|---|---|---|
| OFR-0678SaladCloud | Unspecified, Multiple / Unspecified | NVIDIA GeForce RTX 3080 Ti Reference specification1 GPU configuration · 1 GPU per node | UnconfirmedConfirmed capacity: Not confirmed | $0.08 USD / GPU-hourOriginal: USD 0.08 GPU/hour | On-DemandSalad Container Engine · Lowest priority · Per-second · Lowest priority | Confidence AChecked todayConfirmed 29 Sept 2026Source: SRC000036 | Inspect |
| OFR-0851SaladCloud | Unspecified, Multiple / Unspecified | NVIDIA GeForce RTX 3080 Ti Reference specification1 GPU configuration · 1 GPU per node | UnconfirmedConfirmed capacity: Not confirmed | $0.12 USD / GPU-hourOriginal: USD 0.12 GPU/hour | On-DemandSalad Container Engine · Low priority · Per-second · Low priority | Confidence AChecked todayConfirmed 29 Sept 2026Source: SRC000036 | Inspect |
| OFR-0850SaladCloud | Unspecified, Multiple / Unspecified | NVIDIA GeForce RTX 3080 Ti Reference specification1 GPU configuration · 1 GPU per node | UnconfirmedConfirmed capacity: Not confirmed | $0.16 USD / GPU-hourOriginal: USD 0.16 GPU/hour | On-DemandSalad Container Engine · Medium priority · Per-second · Medium priority | Confidence AChecked todayConfirmed 29 Sept 2026Source: SRC000036 | Inspect |
| OFR-0849SaladCloud | Unspecified, Multiple / Unspecified | NVIDIA GeForce RTX 3080 Ti Reference specification1 GPU configuration · 1 GPU per node | UnconfirmedConfirmed capacity: Not confirmed | $0.20 USD / GPU-hourOriginal: USD 0.20 GPU/hour | On-DemandSalad Container Engine · High priority · Per-second · High priority | Confidence AChecked todayConfirmed 29 Sept 2026Source: SRC000036 | Inspect |
| OFF000393Exoscale | Vienna, Austria | NVIDIA GeForce RTX 3080 Ti Reference specificationMinimum 2 GPU configuration · 2 GPUs per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 1.71 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked todayConfirmed 02 Oct 2026Source: SRC-EXOSCALE-PRICING-API-20261031 | Inspect |
| OFF000394Exoscale | Vienna, Austria | NVIDIA GeForce RTX 3080 Ti Reference specificationMinimum 4 GPU configuration · 4 GPUs per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 3.14 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked todayConfirmed 02 Oct 2026Source: SRC-EXOSCALE-PRICING-API-20261031 | Inspect |
| OFF000392Exoscale | Vienna, Austria | NVIDIA GeForce RTX 3080 Ti Reference specificationMinimum 1 GPU configuration · 1 GPU per node | UnknownConfirmed capacity: Not confirmed | Comparable price unavailableOriginal: EUR 0.92 instance-hour | On-DemandVirtual Machine · Per-second | Confidence BChecked todayConfirmed 02 Oct 2026Source: SRC-EXOSCALE-PRICING-API-20261031 | Inspect |
Recorded hardware attributes. Not tracked is distinct from No.
Public evidence and confidence signals supporting Kovara-tracked coverage.