AI Model Watch
Models / OpenAI

text-embedding-3-large

GA

Most capable embedding model. Output up to 3072 dimensions (reducible via the dimensions param). Price is first-party: $0.13 per 1M input tokens, confirmed unchanged 2026-08-08 against developers.openai.com/api/docs/pricing (Standard tier, embeddings table). Embeddings carry no output-token billing. Max input tokens CORRECTED 8191 -> 8192 on 2026-08-18: the model spec page states no limit at all, and the live API reference (developers.openai.com/api/docs/api-reference/embeddings/create) states it as a constraint on the input parameter -- "the max input tokens for the model (8192 tokens for all embedding models)". 8191 was the figure OpenAI's 2024 new-embedding-models announcement carried; the reference carries the field as data and is current, so it wins on the precision ranking. Not an upstream change and therefore no changelog entry.

text-embedding-3-large by OpenAI is tracked live in our LLM catalog, with a 8K (8192-token) context window. It is generally available (GA).

Last verified: 19 Aug 2026 · sourced from official provider documentation

Provider
OpenAI
Status
GA
Input price
$0.13 / 1M tokens
Output price
—
Cached input
—
Blended price
—
Context window
8192 tokens (8K)
Max output
—
Embedding dimensions
3072-dim output vectors
Modality
text
Open weights
No — API only
Knowledge cutoff
—
Released
25 Jan 2024
API string
text-embedding-3-large

Source: OpenAI official documentation ↗

See also

Explore more

All OpenAI models→

Prices, context windows and lifecycle status for every OpenAI model we track.

Compare models head-to-head→

Side-by-side price, context and status for the models developers weigh against each other.

FAQ

text-embedding-3-large — questions & answers

What is the context window of text-embedding-3-large?

text-embedding-3-large has a 8192-token context window (8K).

Is text-embedding-3-large deprecated?

No — text-embedding-3-large is generally available (GA) and not currently scheduled for deprecation or retirement in our tracker.

Does text-embedding-3-large support image or vision input?

Our catalog lists text-embedding-3-large's modalities as text; image/vision input is not among them.

Can I self-host text-embedding-3-large?

No — text-embedding-3-large is proprietary. Its weights are not publicly released, so it is available only through OpenAI's API (or hosted partners), not for self-hosting.

What is the API model string for text-embedding-3-large?

The API model identifier for text-embedding-3-large is "text-embedding-3-large" when calling OpenAI's API.