Titan Embeddings V2
Embedding
AWS-native text embedding model. Default for Bedrock Knowledge Bases and AWS-aligned RAG.
Titan Embeddings V2 is a embedding AI model from Amazon. It costs $0.020 per million input tokens, with a 8,000-token context window.
INPUT
$0.020/M
per million input tokens
OUTPUT
—
per million output tokens
CONTEXT
8,000
tokens
Benchmarks
No published benchmark scores tracked for this model yet. Frontier and reasoning models from major providers have scores; smaller models and inference-host variants typically inherit the underlying open-weights score.
More from Amazon
See all 10 →Nova 2 Pro
Frontier · 1,000,000 ctx
in $0.800/Mout $3.200/M
Nova 2 Lite
Efficient · 1,000,000 ctx
in $0.300/Mout $2.500/M
Nova Premier
Top-tier · 300,000 ctx
in $2.500/Mout $12.500/M
Nova Pro
Frontier · 300,000 ctx
in $0.800/Mout $3.200/M
Nova Lite
Efficient · 300,000 ctx
in $0.060/Mout $0.240/M
Nova Micro
Micro · 128,000 ctx
in $0.035/Mout $0.140/M
Frequently asked questions
How much does Titan Embeddings V2 cost?
Titan Embeddings V2 costs $0.020 per million input tokens.
What is Titan Embeddings V2's context window?
Titan Embeddings V2 supports up to 8,000 tokens of context in a single request.
What is Titan Embeddings V2 best for?
AWS-native text embedding model. Default for Bedrock Knowledge Bases and AWS-aligned RAG.
Who makes Titan Embeddings V2?
Titan Embeddings V2 is developed and served by Amazon.