Amazon - API - verified
Amazon Nova Multimodal Embeddings embeddings
Amazon Nova Multimodal Embeddings is a current embedding model from Amazon, called by API as amazon.nova-2-multimodal-embeddings-v1:0. It returns vectors of 256, 384, 1,024 and 3,072 dimensions (no default is stated) and accepts up to 8,000 input tokens, for text, image, audio and video input. The first-party list price is $0.135 per 1M input tokens ($0.0675 through the batch API).
CurrentModel id: amazon.nova-2-multimodal-embeddings-v1:0
- Default dims
- Not stated
- Max input
- 8,000 tok
- Price / 1M
- $0.135
- Inputs
- Text, Image, Audio, Video
Dimensions
Flexible output size- 3,072
- 1,024
- 384
- 256
Default output: not stated by the provider.
Listed sizes: 3,072, 1,024, 384 and 256.
Amazon documents shorter output vectors for this model (Matryoshka-style), so you can store smaller embeddings by asking for fewer dimensions instead of re-embedding with another model.
Pricing
- Per 1M input tokens
- $0.135
- Batch, per 1M
- $0.0675
Text input price; images $0.00006 each (document images $0.0006), video $0.0007 per second and audio $0.00014 per second on demand.
Source: aws.amazon.com/bedrock/pricing
Usage
Copied verbatim from the page linked below. Placeholders such as project ids and API keys are the source's own.
# Define the request body.
request_body = {
'taskType': 'SINGLE_EMBEDDING',
'singleEmbeddingParams': {
'embeddingPurpose': 'GENERIC_INDEX',
'embeddingDimension': 3072,
'text': {'truncationMode': 'END', 'value': 'Hello, World!'},
},
}
try:
# Invoke the Nova Embeddings model.
response = bedrock_runtime.invoke_model(
body=json.dumps(request_body, indent=2),
modelId='amazon.nova-2-multimodal-embeddings-v1:0',
accept='application/json',
contentType='application/json',
)Source: docs.aws.amazon.com/nova/latest/nova2-userguide/embeddings.html
Worth knowing
Maps text, document images, images, video and audio into one space for cross-modal retrieval and RAG, with a maximum context of 8K tokens or 30s of video or audio and async segmentation for long inputs.
Released . Multilingual: not stated.
FAQ
How many dimensions does Amazon Nova Multimodal Embeddings output?
Default output: not stated by the provider. Listed sizes: 3,072, 1,024, 384 and 256. Amazon documents shorter output vectors for this model (Matryoshka-style), so you can store smaller embeddings by asking for fewer dimensions instead of re-embedding with another model.
What is the max input length of Amazon Nova Multimodal Embeddings?
8,000 tokens per input, per Amazon's documentation. Longer text has to be chunked before you embed it.
How much does Amazon Nova Multimodal Embeddings cost?
The first-party list price is $0.135 per 1M input tokens ($0.0675 through the batch API). Text input price; images $0.00006 each (document images $0.0006), video $0.0007 per second and audio $0.00014 per second on demand. Source: https://aws.amazon.com/bedrock/pricing/.
What can Amazon Nova Multimodal Embeddings embed?
Text, image, audio and video input, per the provider.
Sources
Last verified from the pages below.
- Amazon Nova 2 embeddings guide - docs.aws.amazon.com
- Bedrock model card - docs.aws.amazon.com
- Amazon Bedrock pricing - aws.amazon.com
More Amazon embedding models
Default dimensions, max input tokens and list price per 1M tokens.
Newsletter
New models, every week
One email per week. New model releases, embedding and retrieval updates and what they mean for what you build.
Keep exploring
Model API pricing and specs
First-party prices per 1M tokens for chat and reasoning models.
Local models
Open-weight LLMs you can download and run, with sizes and licenses.
What is RAG?
Where embeddings fit in a retrieval pipeline, end to end.
Vector databases compared
Where the vectors go once you have picked a model.