BedrockExplorer

Embed v4

Cohere · Active · available in 23 of 31 regions

Cost calculator Compare
VisionEmbeddings Tool usePrompt cachingBatch inferenceStreamingFine-tuning
Context window
128K
128,000 tokens
Max output
—
Input / 1M
—
Output / 1M
—

What Embed v4 is good for

editorial

An embedding model: it turns text into vectors for search and clustering, and does not generate text.

Suits

  • Retrieval. Indexing a corpus for semantic search or deduplication.
  • Chunk size. Accepts up to 128K tokens per call, which sets how large your chunks can be.

Think twice if

  • Not for generation. It cannot answer questions or write text — pair it with a text model.

This section is judgement, not data from AWS. It is composed from the capabilities, context window, price position and region coverage shown elsewhere on this page — so it stays in step with the daily snapshot rather than going stale.

Where teams typically use it

Embeddings

Turning text into vectors for search, clustering and deduplication. Priced per input token, with no output cost.

  • · Semantic search across a knowledge base where keyword search keeps missing synonyms.
  • · Deduplicating a CRM by finding records that describe the same company under different names.
  • · Clustering support tickets to see which themes are actually growing.

Model & inference profile IDs

Base model ID — direct on-demand invoke

Cross-region inference profiles (3)

Regions marked Inference profile only require a profile ID rather than the base model ID — invoking the base ID there returns a validation error. Regional (us., eu.) profiles carry a 10% premium over global.

Pricing by region

USD per 1M tokens

No published token pricing

The AWS Price List API publishes no token SKU for this model, and it is not in our curated overlay. This is common for embedding, image and video models, which bill per request or per second rather than per token.

Check the AWS pricing page →

Region availability

On-demand us-east-1 On-demand Inference profile only us-east-2 Profile only Inference profile only us-west-1 Profile only Inference profile only us-west-2 Profile only Inference profile only ca-central-1 Profile only Not available ca-west-1 Not available Not available mx-central-1 Not available On-demand eu-west-1 On-demand Inference profile only eu-west-2 Profile only Inference profile only eu-west-3 Profile only Inference profile only eu-central-1 Profile only Inference profile only eu-central-2 Profile only Inference profile only eu-north-1 Profile only Inference profile only eu-south-1 Profile only Inference profile only eu-south-2 Profile only Not available ap-east-2 Not available On-demand ap-northeast-1 On-demand Inference profile only ap-northeast-2 Profile only Inference profile only ap-northeast-3 Profile only Inference profile only ap-south-1 Profile only Inference profile only ap-south-2 Profile only Inference profile only ap-southeast-1 Profile only Inference profile only ap-southeast-2 Profile only Inference profile only ap-southeast-3 Profile only Inference profile only ap-southeast-4 Profile only Not available ap-southeast-5 Not available Not available ap-southeast-6 Not available Not available ap-southeast-7 Not available Not available il-central-1 Not available Not available af-south-1 Not available Inference profile only sa-east-1 Profile only

TPM / RPM are default account quotas from AWS Service Quotas where a model-specific limit is published. They are per-account defaults and adjustable on request.

Where this model actually runs

Under a strict residency constraint, Embed v4 can be served without the request leaving 4 of the 25 jurisdictions with a Bedrock region.

EU Ireland Japan United States All jurisdictions →

Embed v4 appears in

Embedding models

Common questions

Which AWS regions support Embed v4?
23 of 31 regions: us-east-1, us-east-2, us-west-1, us-west-2, ca-central-1, eu-west-1, eu-west-2, eu-west-3, eu-central-1, eu-central-2, eu-north-1, eu-south-1, eu-south-2, ap-northeast-1, ap-northeast-2, ap-northeast-3, ap-south-1, ap-south-2, ap-southeast-1, ap-southeast-2, ap-southeast-3, ap-southeast-4, sa-east-1. Regions marked profile-only need a cross-region inference profile ID such as eu.cohere.embed-v4:0 rather than the bare model ID.
What is the context window of Embed v4?
128,000 tokens (128K).
What is the model ID for Embed v4 on Bedrock?
cohere.embed-v4:0. In regions where it is only reachable through a cross-region inference profile, use a prefixed ID instead: eu.cohere.embed-v4:0, global.cohere.embed-v4:0, us.cohere.embed-v4:0.

Other Cohere models

Embed English Embed Multilingual Rerank 3.5 Compare these →