
5 offers at spot / preemptible rates across 2 providers that publish this card — cheapest first, in USD (non-USD marked “≈”). Rates are per GPU-hour; the total is what the whole configuration costs.
"[ALT Config]" / "[Premium]" / "NVLink" suffixes seen on Massed Compute describe the surrounding host config, not a different chip — see providers.yaml notes for that source. Do not split these into separate canonical models.
Verdaformerly DataCrunchno comparable on-demand rate
Hyperstackno comparable on-demand rate