OpenAI: text-embedding-ada-002

Text Embedding Ada 002 is an earlier OpenAI embedding model used in many existing vector indexes. It remains relevant when maintaining compatibility with stored embeddings, since vectors from different models should not be mixed in one similarity space.

Compare models

text-embedding-ada-002

AvailableOpenAIEmbeddingSync

Input / Output: $0.10 / $0.00
Context: 8.2K
Modalities: Embedding

About text-embedding-ada-002

text-embedding-ada-002 is an earlier OpenAI embedding model that many existing vector indexes were built on. OpenAI describes it as an improved, more performant version of the original ada embedding model, covering search, clustering, recommendations, anomaly detection, and classification. It is mainly relevant for staying compatible with stored vectors, since the newer text-embedding-3 models replace it for new projects.

Where it works well

Compatible with the large number of indexes already built with ada-002 vectors.
Covers search, clustering, recommendations, anomaly detection, and classification.
A single, familiar embedding model with a long record in production systems.
Works on the embeddings and batch endpoints.

When to choose another model

OpenAI's text-embedding-3 models are the newer generation, and new projects usually start there.
Its vectors cannot be mixed with another model's, so migrating means re-embedding everything.
It embeds text only and offers no selectable dimensions.

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-ada-002",
  "input": "The quick brown fox jumps over the lazy dog."
}'

Parameters

Embedding

Parameter	Values and limits	Default	Notes
`encoding_format`	float, base64	Not provided	Not provided
`input`	Inputs: ≥ 1, ≤ 2048	Not provided	Input accepts text or token IDs. Each input is limited to 8192 tokens; a request is limited to 300000 tokens in total.
`user`	Not provided	Not provided	Not provided

Pricing

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

Embedding

per 1M tokens

Official price: Input $0.10 / Output $0.00
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Embedding	Input $0.10 / Output $0.00	—	—

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 Nov 1, 2026, 10:22 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-ada-002 in Console with a prompt ready to edit or send.

Help me try text-embedding-ada-002 with embedding at /v1/embeddings. Show the result and cost.

Use cases

Maintaining an existing index
Add new documents to a corpus already embedded with ada-002 so queries still match.
Legacy RAG systems
Keep a production retrieval stack stable and unchanged while the team plans a careful migration.
Reproducing prior results
Re-run older experiments and evaluations that depended on ada-002 vectors, so results stay comparable.
Migration baseline
Compare retrieval quality of ada-002 against a newer embedding model on your own queries before switching.

Prompt examples

Embed these new support articles so they match the existing ada-002 index.

Embed my test queries with the same model as the stored corpus.

Embed 500 documents so I can compare retrieval against text-embedding-3-small.

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-ada-002?

It is an older OpenAI text embedding model that OpenAI describes as an improved, more performant version of the original ada embedding model. It turns text into vectors for search and similar tasks.

Why would anyone still use it?

To stay compatible with vectors already stored. A query must be embedded with the same model as the documents, so an existing ada-002 index needs ada-002 until you re-embed.

Should new projects use it?

Usually not. The text-embedding-3 models are newer and OpenAI lists text-embedding-3-small as an improvement over the ada model. Start there unless compatibility with stored vectors forces ada-002.

How do I migrate away from it?

Re-embed the full corpus with the new model, build a new index, and switch queries over together. Never mix vector spaces in one index, and keep the old index until the new one is verified.

Which endpoints support it?

It works with the embeddings endpoint and the batch endpoint only. Chat, assistants, fine-tuning, and image endpoints are not supported, so it serves purely as a vector generator for retrieval systems.

How much does text-embedding-ada-002 cost?

On TokenLab, text-embedding-ada-002 costs Input $0.10 / 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-ada-002 use?

Use https://api.tokenlab.sh/v1/embeddings for text-embedding-ada-002. The request example below shows the matching code shape.

Which operations does text-embedding-ada-002 support?

text-embedding-ada-002 supports Embedding. Select an operation above to see its endpoint and request example.