Praison AI

AWS Bedrock Embeddings

Generate embeddings using Amazon Bedrock

Overview

Amazon Bedrock provides access to embedding models from Amazon Titan, Cohere, and other providers through AWS infrastructure.

Quick Start

python
from praisonaiagents import embedding

result = embedding(
    input="Hello world",
    model="bedrock/amazon.titan-embed-text-v1"
)
print(f"Dimensions: {len(result.embeddings[0])}")

CLI Usage

bash
praisonai embed "Hello world" --model bedrock/amazon.titan-embed-text-v1

Setup

bash
export AWS_ACCESS_KEY_ID="your-access-key"
export AWS_SECRET_ACCESS_KEY="your-secret-key"
export AWS_REGION_NAME="us-east-1"

Available Models

ModelDimensionsProvider
bedrock/amazon.titan-embed-text-v11536Amazon
bedrock/amazon.titan-embed-text-v2:01024Amazon
bedrock/amazon.nova-2-multimodal-embeddings-v1:01024Amazon
bedrock/cohere.embed-english-v31024Cohere
bedrock/cohere.embed-multilingual-v31024Cohere
bedrock/cohere.embed-v4:01024Cohere

With AWS Profile

python
from praisonaiagents import embedding
import os

os.environ["AWS_PROFILE"] = "my-profile"

result = embedding(
    input="Hello world",
    model="bedrock/amazon.titan-embed-text-v1"
)

Batch Embeddings

python
from praisonaiagents import embedding

texts = ["Document 1", "Document 2", "Document 3"]
result = embedding(
    input=texts,
    model="bedrock/amazon.titan-embed-text-v1"
)
print(f"Generated {len(result.embeddings)} embeddings")

Cross-Region Usage

python
from praisonaiagents import embedding

result = embedding(
    input="Hello world",
    model="bedrock/amazon.titan-embed-text-v1",
    aws_region_name="eu-west-1"
)

Related