Answer: The newest and most performant embedding models are text-embedding-3-small and text-embedding-3-large.
Status: Confirmed
Entry: model/openai-embedding
Revision: 1
Digest: sha256:c808cc7a9217eb61 (changes whenever the content changes)

Timeline:
  Created: 2026-11-12T10:44:25Z
  Last changed: 2026-11-12T10:44:25Z
  Last verified: 2026-11-12T11:34:02Z
  Next check: 2026-11-13T11:34:02Z
  Check interval: Daily
  Expected change: No fixed schedule

Data:
  Field | Value | Unit | Source | As of
  newest embedding models | text-embedding-3-small and text-embedding-3-large |  | [1] | 2026-11-12
  generation | third-generation |  | [1] | 2026-11-12
  default dimensions (small) | 1536 | dimensions | [1] | 2026-11-12
  default dimensions (large) | 3072 | dimensions | [1] | 2026-11-12
  MTEB performance (small) | 62.3% |  | [1] | 2026-11-12
  MTEB performance (large) | 64.6% |  | [1] | 2026-11-12

Sources:
  [1] OpenAI. "platform.openai.com/docs/guides/embeddings". https://platform.openai.com/docs/guides/embeddings
      Role: primary. Accessed: 2026-11-12T11:34:02Z
      Quote: "New embedding models text-embedding-3-small and text-embedding-3-large , our newest and most performant embedding models, are now available." (Supports: newest embedding models)
      Quote: "OpenAI offers two powerful third-generation embedding model (denoted by -3 in the model ID)." (Supports: generation)
      Quote: "By default, the length of the embedding vector is 1536 for text-embedding-3-small or 3072 for text-embedding-3-large ." (Supports: default dimensions (small), default dimensions (large))
      Quote: "text-embedding-3-small 62,500 62.3% 8192" (Supports: MTEB performance (small))
      Quote: "text-embedding-3-large 9,615 64.6% 8192" (Supports: MTEB performance (large))
      Published: 2026-11-12T11:14:48Z
      Snapshot SHA-256: daf663929671fccf415de3e809052cf620112e8009b44f0903c31f4bba513db2 (captured 2026-11-12T10:44:25Z)

Verification:
  Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-12T10:44:25Z, so the values verified then were kept.
  Cross-check: not-available at 2026-11-12T11:34:02Z. No independent second source is configured for this entry
  Last successful check: 2026-11-12T11:34:02Z
  Last attempt: 2026-11-12T11:34:02Z

Notes and limitations:
  Extracted automatically from a single official source. Only the first 60,000 characters of the page text were read.

Change log:
  Time | Event | Change | Source
  2026-11-12T10:44:25Z | added | Entry created | —
  Changes for this entry as a feed: https://latestn.com/model/openai-embedding/latest

Challenge this entry:
  Open challenges: None
  How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so.
  Form: https://latestn.com/dispute/model/openai-embedding
  POST: https://latestn.com/dispute/model/openai-embedding with claim, evidence_url, and optionally field and contact (form-encoded or JSON)
  GET: https://latestn.com/dispute/model/openai-embedding?claim={what is wrong}&evidence_url={link}

How to cite:
    latestN. (2026). What is the latest OpenAI embedding model? Revision 1, last verified 2026-11-12T11:34:02Z. https://latestn.com/model/openai-embedding (accessed 2026-11-12T12:52:44Z).

    @misc{latestN_model_openai_embedding_r1,
      title        = {What is the latest OpenAI embedding model?},
      author       = {{latestN}},
      year         = {2026},
      howpublished = {\url{https://latestn.com/model/openai-embedding}},
      note         = {Revision 1; digest sha256:c808cc7a9217eb61; last verified 2026-11-12T11:34:02Z; accessed 2026-11-12T12:52:44Z}
    }

URL: https://latestn.com/model/openai-embedding