# Amazon Nova Multimodal Embeddings (slim)

> Amazon Nova Multimodal Embeddings is an AWS model on Amazon Bedrock that turns text, images, document images, video and audio into vectors in one space, at 256, 384, 1024 or 3072 dimensions, through synchronous and asynchronous calls.

- Full: https://www.anchorterminal.com/tools/amazon-nova-embeddings.md (~8,050 tokens) · this version ~1,480 tokens · JSON https://www.anchorterminal.com/tools/amazon-nova-embeddings.json · canonical https://www.anchorterminal.com/tools/amazon-nova-embeddings
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-09

**BB · 75/100 · rank #59 of 842 · #1 in Embeddings & rerankers · agent-ready · confidence medium**


Assessment: One model embeds text, images, document images, video and audio into a shared space, with nine documented purpose settings and published per-unit prices. It runs in US East (N. Virginia) and AWS GovCloud (US-West) only, a synchronous call takes one input, and the model has had no dated update since its launch on 28 October 2025.

## Facts

- Kind: HTTP API · vendor: Amazon Web Services · category: Embeddings & rerankers · legal entity: Amazon Web Services, Inc. · provenance 88/100
- Endpoint: `https://bedrock-runtime.us-east-1.amazonaws.com` (HTTP)
- Auth: API key · pricing: Pay per use · x402: no · licence: Proprietary service under the AWS Service Terms. The AWS SDKs are Apache-2.0
- Probe metrics: not measured yet (probes haven't run)
- Model ID: `amazon.nova-2-multimodal-embeddings-v1:0`
- Endpoint: `https://bedrock-runtime.{region}.amazonaws.com`. Not served on the `bedrock-mantle` endpoint
- Regions: us-east-1 (N. Virginia) and us-gov-west-1 (GovCloud), in-Region only
- Dimensions: 256, 384, 1024 or 3072. Default 3072
- Modalities: Text, image (standard or document detail), audio and video
- Synchronous limits: One input per request. 8,192 characters of inline text, 30 seconds of audio, 30 seconds of video. AWS states a context of 8K tokens
- Asynchronous segmentation: Text segments of 800 to 50,000 characters (default 32,000). Audio and video segments of 1 to 30 seconds (default 5)
- Formats: Images png, jpeg, gif, webp. Audio mp3, wav, ogg. Video mp4, mov, mkv, webm, flv, mpeg, mpg, wmv, 3gp
- Embedding purposes: `GENERIC_INDEX`, `GENERIC_RETRIEVAL`, `TEXT_RETRIEVAL`, `IMAGE_RETRIEVAL`, `VIDEO_RETRIEVAL`, `DOCUMENT_RETRIEVAL`, `AUDIO_RETRIEVAL`, `CLASSIFICATION`, `CLUSTERING`
- Quotas: 2,000 requests a minute on demand, 30 concurrent asynchronous requests, batch jobs of 100 to 100,000 records
- Data retention: Bedrock states zero data retention by default. This model is not among those listed for abuse-detection storage. Image inputs flagged as apparent CSAM may be stored and reviewed
- Lifecycle: Active. End of life no sooner than 28 October 2026, with a Legacy period of at least 6 months before any end of life
- Reranker: None in this model
- Prices: Text input, on demand $0.135 per 1M tokens; Text input, batch $0.0675 per 1M tokens; Standard image input $0.0001 per image; Document image input $0.0006 per image; Video input $0.0007 per second of video; Audio input $0.0084 per minute of audio

- Scores: Reliability 95, Performance pending, Schema & documentation 76, Agent ergonomics 78, Security & auth 91, Payments & pricing 30, Task success pending, Maintenance & community 50, Transparency & trust 79 · total over the 7 assessed categories
- Why: Reliability, Hosted reading. · Schema & documentation, Model reading. · Agent ergonomics, Model reading. · Security & auth, Model reading. · Payments & pricing, No x402, MPP or L402 (0). · Maintenance & community, Model reading. · Transparency & trust, Closed model under the AWS Service Terms, with Apache-2.0 SDKs (15).
- Sources: 32, open questions: 7, both in the full twin
- Capabilities: embed.text, embed.multimodal
- JSON: https://www.anchorterminal.com/api/v1/tools/amazon-nova-embeddings.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/amazon-nova-embeddings.svg` or a link to https://www.anchorterminal.com/tools/amazon-nova-embeddings from a page on amazon.com or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Call `bedrock-runtime` in us-east-1 with model ID `amazon.nova-2-multimodal-embeddings-v1:0`. No other commercial Region serves it
2. Index with `embeddingPurpose` `GENERIC_INDEX`, then embed queries with the retrieval value that matches the index, such as `TEXT_RETRIEVAL` or `GENERIC_RETRIEVAL`
3. Always send `truncationMode` with text. It is required, and `NONE` fails the request when the text is too long
4. Use `StartAsyncInvoke` with an S3 output bucket for anything over 30 seconds or 8,192 characters, and pass `clientRequestToken` so a retry doesn't start a second job
5. Keep one `embeddingDimension` per index. The default is 3072

## Connect

```bash
pip install boto3
```
Full config and headless snippets are in the full page. Through letme (picks today, calling later): https://letme.dev/amazon-nova-embeddings

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Cohere Embed and Rerank | BB | 72.5 | embed.text, embed.multimodal | https://www.anchorterminal.com/tools/cohere-embed.min.md |
| Gemini Embedding | BB | 70.6 | embed.text, embed.multimodal | https://www.anchorterminal.com/tools/gemini-embedding.min.md |
| Jina Embeddings and Reranker | C | 61 | embed.text, embed.multimodal | https://www.anchorterminal.com/tools/jina-embeddings.min.md |
| NVIDIA NeMo Retriever Embedding and Reranking NIMs | C | 61 | embed.text, embed.multimodal | https://www.anchorterminal.com/tools/nvidia-nemo-retriever.min.md |
| Voyage AI embeddings and rerankers | C | 58.8 | embed.text, embed.multimodal | https://www.anchorterminal.com/tools/voyage-ai.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)