Plain-English Summary
An AWS embedding model that places text, images, audio, video, and documents into a shared vector space.
Amazon Nova Multimodal Embeddings
An AWS embedding model that places text, images, audio, video, and documents into a shared vector space.
Nova 2
Embedding
GA
Open weights: No
API: Yes — Amazon Bedrock
Technical Notes
Multimodal embedding model for cross-modal retrieval, semantic search, and recommendation systems.
Best For
Multimodal search, media libraries, RAG, recommendations
Watch Out For
Requires AWS and a vector index; application quality depends on domain evaluation.
Model Specs
- Maker
- Amazon Web Services
- Country
- Not listed
- Family
- Nova 2
- Type
- Embedding
- Release date
- 2025-12-02
- Date confidence
- Exact
- Status
- GA
- Context window
- Varies
- Max output
- Not listed
- Parameters
- Not disclosed
- Active parameters
- Not disclosed
- Architecture
- Proprietary multimodal embedding model
- Reasoning
- Not applicable
- Tool calling
- No
- Structured output
- Embedding vector
- API available
- Yes — Amazon Bedrock
- Open weights
- No
- Self-hostable
- No
- License
- Proprietary
- Inputs
- Text, image, audio, video, documents
- Outputs
- Embedding vector
- Approx. price
- See provider pricing
- How to access
- Provider website
- Model / API ID
- amazon.nova-multimodal-embeddings
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Vector dimensions and AWS retrieval stack
- Completeness
- 81 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source