Cohere

Embed v4.0 API Pricing

Configurable output dimensions (256-1536). Only Model Vault dedicated-instance pricing ($4-5/hr or $2,500-3,250/mo) was published; no public self-serve per-token rate found as of verification date.

Could not verify pricing on 2026-07-02 — treat these numbers as tentative.

View source ↗

Facts

Unverified

Specs

  • FamilyEmbed
  • API IDembed-v4.0
  • ModalityEmbedding
  • Context window128K tokens
  • Max output
  • Release dateNot published

Pricing (per 1M tokens)

  • Input
  • Output
  • Cached input
  • Cache write
  • Batch discount

Capabilities

embedding, multimodal

Notes

Configurable output dimensions (256-1536). Only Model Vault dedicated-instance pricing ($4-5/hr or $2,500-3,250/mo) was published; no public self-serve per-token rate found as of verification date.

Worked cost examples

Not priced per token

Embed v4.0is billed per unit (image, second, or similar), not per token — the standard chatbot/RAG/batch workloads don’t apply. See the notes in the facts panel above for its per-unit rate.

Price history

Changes we have documented since we began tracking

No price changes recorded since we began tracking.

Embed v4.0 pricing FAQ

How is Embed v4.0 priced?
Embed v4.0 is billed per unit rather than per token. Configurable output dimensions (256-1536). Only Model Vault dedicated-instance pricing ($4-5/hr or $2,500-3,250/mo) was published; no public self-serve per-token rate found as of verification date.
Does Embed v4.0 have cache pricing?
No cache discount is documented for Embed v4.0 on the provider's pricing page.