text-embedding-3-small

text-embedding-3-small is the inexpensive half of OpenAI’s embedding pair. It produces 1,536 numbers per text — the same size as the older ada-002, so it drops into an existing search index — and beats ada-002 on both the MTEB benchmark (62.3% against 61.0%) and multilingual search.

For English-language search and recommendations it is close enough to the large model that the saving usually wins. Where documents are in several languages, the large model’s lead is wide enough to be worth paying for.

Every rate is published and stays the same from one request to the next. See every model.

text-embedding-3-small is billed in rupees, on the same monthly invoice as every other model your team uses — no foreign card, no separate OpenAI account, and one key for all of it.

Prompts sent to text-embedding-3-small are processed outside Sri Lanka. If a workload has to stay in the country, Qwen3.8-27B (Colombo) runs on our own servers in Colombo.

Use the OpenAI SDK you already have — point it at Roar AI and name the model. Read the docs.

from openai import OpenAI

client = OpenAI(base_url="https://api.roar-ai.com/v1", api_key="roar_live_…")

vectors = client.embeddings.create(
    model="text-embedding-3-small",
    input=["Quarterly sales report, western province"],
)