Amazon Nova Multimodal Embeddings vs OpenAI embeddings

Amazon Nova Multimodal Embeddings

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.

OpenAI embeddings

text-embedding-3-small at $0.02 per million tokens, $0.01 through the Batch API. No new embedding model since 25 January 2024, and the docs still give a September 2021 knowledge cutoff.

Amazon Nova Multimodal Embeddings

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

OpenAI embeddings

1. Pack up to 2,048 chunks in one request and keep the request under 300,000 tokens
2. Count tokens before sending. An input over 8,192 tokens is rejected, not truncated
3. Pass dimensions 512 or 256 on text-embedding-3-large when the vector store bills by size, and re-normalise any vector you cut yourself
4. Split a Batch API index job into batches of under 50,000 inputs. It's half price with a 24-hour window
5. Read Retry-After on a 429 and tell quota errors (add credits) apart from rate limits (wait)