Fact · Voice and AI
0.526US dollars per hour
The AWS g4dn.xlarge instance, which has one NVIDIA T4 GPU, starts at $0.526 per hour on demand in the US East (N. Virginia) region, according to a price tracker updated October 16, 2026.
Checked against its source on October 17, 2026
In the source's own words
“The g4dn.xlarge is an Accelerated computing instance with 4 vCPUs and 16 GiB of memory, 1x NVIDIA T4 with up to 5 Gbps bandwidth, starting at $0.526 per hour.”
AWS g4dn.xlarge Price (on-demand, us-east-1), Holori, 2026, Price section.
Open the original- What it measures
- A self-hosted speech model runs on a rented server that bills for every hour it is switched on, whether or not anyone is dictating. The tracker lists this price for an instance with a single GPU, a common size for running a speech model.
- What it does not show
- The price is from a third-party tracker. Reserved and spot pricing are cheaper, and other regions cost more. The instance is only one choice; a smaller or larger GPU changes the price.