NVIDIA Nemotron Nano 12B v2 VL BF16
LegacyNVIDIA · on Amazon Bedrock · last checked Oct 19, 2026
AWS card says: “Legacy”, Model EOL date: March 30, 2027. AWS does not name a specific replacement model on its lifecycle page or model card. lifecycle page
Active NVIDIA models on Bedrock, newest first (a list, not an AWS recommendation): NVIDIA Nemotron 3 Super 120B, Nemotron Nano 3 30B.
Model IDs and inference profile IDs
| Endpoint | Model ID | Geo profile IDs | Global profile ID |
|---|---|---|---|
| bedrock-runtime | nvidia.nemotron-nano-12b-v2https://bedrock-runtime.{region}.amazonaws.com | us-gov.nvidia.nemotron-nano-12b-v2 | none |
| bedrock-mantle | nvidia.nemotron-nano-12b-v2https://bedrock-mantle.{region}.api.aws/v1 | none | none |
Lifecycle
- Status
- Legacy
- Launched
- Oct 28, 2025
- EOL no sooner than
- Oct 28, 2026
- EOL date
- Mar 30, 2027
- Legacy period
- at least 6 months
- Context window
- 128K tokens
- Max output
- 8K
- Input
- Image, Text
- Output
- Text
Retirement notices
| Regions | Legacy from | Extended access from | EOL |
|---|---|---|---|
| ap-northeast-1, ap-south-1, ap-southeast-2, eu-south-1, eu-west-1, eu-west-2, sa-east-1, us-east-1, us-east-2, us-west-2 | Sep 30, 2026 | To be announced | Mar 30, 2027 |
During public extended access the provider may charge more. AWS lifecycle page
Regions and on-demand prices
USD per 1M tokens, standard on-demand tier (no batch, cache, flex, priority or long-context rates). From the AWS Price List API, version published Oct 16, 2026 / Oct 12, 2026.
| Region | Availability |
|---|---|
| us-east-1 N. Virginia | In-Region |
| us-east-2 Ohio | In-Region |
| us-west-2 Oregon | In-Region |
| eu-south-1 Milan | In-Region |
| eu-west-1 Ireland | In-Region |
| eu-west-2 London | In-Region |
| ap-northeast-1 Tokyo | In-Region |
| ap-south-1 Mumbai | In-Region |
| ap-southeast-2 Sydney | In-Region |
| sa-east-1 São Paulo | In-Region |
| us-gov-west-1 AWS GovCloud (US-West) | In-Region |
| us-gov-east-1 AWS GovCloud (US-East) | Inference profile only |
No matching on-demand token price found in the AWS Price List API for this model (image, video, embedding and rerank models are often priced per image, second or request). Check the Amazon Bedrock pricing page.