OpenAI: text-embedding-3-small

Text Embedding 3 Small provides compact vectors for text similarity and retrieval. It is a useful starting point for a search index, recommendation feature, or RAG pipeline that needs an economical embedding model.

Compare models

text-embedding-3-small

AvailableOpenAIEmbeddingSync

Input / Output-30%: $0.02 / $0.00$0.014 / $0.00
Context: 8.2K
Modalities: Embedding

About text-embedding-3-small

text-embedding-3-small is OpenAI's compact embedding model, producing vectors that measure how related two pieces of text are. OpenAI presents it as an improvement over the earlier ada embedding model and lists search, clustering, recommendations, anomaly detection, and classification as uses. It is the economical choice next to text-embedding-3-large, suited to large indexes and RAG pipelines.

Where it works well

An improvement over the ada embedding model, according to OpenAI's model page.
Smaller vectors than the large model keep storage and search light for big indexes.
Covers search, clustering, recommendations, anomaly detection, and classification tasks.
Works on the embeddings and batch endpoints, which fits bulk indexing.

When to choose another model

Retrieval quality is lower than text-embedding-3-large, which matters when answers hinge on precise matches.
It embeds text only, not images or audio.
Switching models later means re-embedding the whole corpus.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

EmbeddingTokenLab endpoint

POST/v1/embeddings

curl -X POST "https://api.tokenlab.sh/v1/embeddings" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "encoding_format": "float",
  "model": "text-embedding-3-small",
  "input": "The quick brown fox jumps over the lazy dog."
}'

Pricing

TokenLab price applies to Verified, which costs less on most models. Official price is the model maker's published price and applies to the more reliable Official route. Auto charges the route that completes the request.

Embedding

per 1M tokens

Official price: Input $0.02 / Output $0.00
TokenLab price: Input $0.014 / Output $0.00
Discount: -30%

	Official priceper 1M tokens	TokenLab priceper 1M tokens	Discount
Embedding	Input $0.02 / Output $0.00	Input $0.014 / Output $0.00	-30%

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 9, 2026, 12:16 AM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open text-embedding-3-small in Console with a prompt ready to edit or send.

Help me try text-embedding-3-small with embedding at /v1/embeddings. Show the result and cost.

Use cases

Default RAG index
Start a retrieval pipeline with an economical embedding model and measure before upgrading.
Recommendations
Find related articles or products by comparing vectors and suggesting the nearest neighbors to each reader.
Anomaly detection
Flag log lines or records that sit far from the cluster of normal ones.
Bulk batch embedding
Embed millions of short records through the batch endpoint when the job can run in the background.

Prompt examples

Embed each FAQ entry so a chatbot can fetch the closest answer.

Embed these product titles and find five similar items for each.

Embed this month's log messages so outliers can be flagged.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What kind of model is text-embedding-3-small?

It is a compact OpenAI embedding model that converts text into numeric vectors. OpenAI calls it an improvement over the previous ada embedding model, and it supports search, clustering, recommendations, anomaly detection, and classification.

When should I choose it over text-embedding-3-large?

Choose small when cost, storage, and throughput matter and retrieval is already adequate. Choose large when quality on difficult or multilingual queries justifies bigger vectors.

Is it better than text-embedding-ada-002?

OpenAI describes it as an improvement over the ada model. If you have an ada-002 index, you must re-embed the corpus to switch, because vectors from different models are not compatible.

Which endpoints work with it?

Two routes: embeddings and batch. Chat, audio, and image endpoints are not supported, so pair it with a separate generation model if you need answers rather than vectors.

Does it take images?

No, it accepts text input only. Use a multimodal embedding model for pictures, and keep this one for text corpora where storage and cost per vector are the main constraints.

How much does text-embedding-3-small cost?

On TokenLab, text-embedding-3-small costs Input $0.014 / Output $0.00 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should text-embedding-3-small use?

Use https://api.tokenlab.sh/v1/embeddings for text-embedding-3-small. The request example below shows the matching code shape.

Which operations does text-embedding-3-small support?

text-embedding-3-small supports Embedding. Select an operation above to see its endpoint and request example.