Which NVIDIA model is cheapest on Amazon Bedrock?

Nemotron Nano 3 30B at $0.060 per 1M input tokens in us-east-1, against $0.240 per 1M output tokens.

Which NVIDIA model is available in the most AWS regions?

NVIDIA Nemotron 3 Super 120B A12B, callable in 11 of the 16 regions tracked here — 11 on-demand and 0 through a cross-region inference profile.

4 NVIDIA models available through Bedrock, with price per 1M tokens, context window and region coverage.

Cheapest

$0.060

per 1M input tokens

Widest availability

of 16 regions

Model	Context	Input	Output	Regions
Nemotron Nano 3 30B nvidia.nemotron-nano-3-30b	—	$0.060	$0.240	On-demand 9 Inference profile only 0
NVIDIA Nemotron 3 Super 120B A12B nvidia.nemotron-super-3-120b	—	$0.150	$0.650	On-demand 11 Inference profile only 0
NVIDIA Nemotron Nano 12B v2 VL BF16 nvidia.nemotron-nano-12b-v2	—	—	—	On-demand 9 Inference profile only 0
NVIDIA Nemotron Nano 9B v2 nvidia.nemotron-nano-9b-v2	—	$0.060	$0.230	On-demand 9 Inference profile only 0

Which NVIDIA model is cheapest on Amazon Bedrock?: Nemotron Nano 3 30B at $0.060 per 1M input tokens in us-east-1, against $0.240 per 1M output tokens.
Which NVIDIA model is available in the most AWS regions?: NVIDIA Nemotron 3 Super 120B A12B, callable in 11 of the 16 regions tracked here — 11 on-demand and 0 through a cross-region inference profile.