by IntfloatReleased 18 November 2025
Multilingual E5 Large turns text in more than 90 languages into a fingerprint of its meaning, so a question in one language can find a matching passage in another. Use it for search across multilingual libraries and for comparing texts in different languages, with each input up to about half a page. Start questions with "query: " and stored passages with "passage: ". For English-only collections, E5 Large v2 is the English-only model with the same level of detail.
You pay for what each request uses. No subscription, no minimum.
One page, the ordinary use
Every rate
| Meter | ↑ you send | ↓ comes back |
|---|---|---|
| Text tokens | 3,33 DA |
per 1M tokens
One company runs this model. OpenDunes sends every request to it. DA per 1M tokens: ↑ what you send.
You pay the price of the provider that served your request. Prices read 15:18 · speed over the last 30 minutes · uptime over 3 days.
Swap the base URL and the key. Your OpenAI code runs unchanged.
import os
from openai import OpenAI
client = OpenAI(
base_url="https://opendunes.com/v1",
api_key=os.environ["OPENDUNES_API_KEY"],
)
resp = client.embeddings.create(
model="intfloat/multilingual-e5-large",
input="Delivery to Oran takes two days.",
)
print(resp.data[0].embedding[:5])Runs this request once on your balance.
About 0,0033 DA per page. ↑ 3,33 DA per 1M tokens.
No. You pay per use from your balance, with no subscription. Run it once from Call it in code, at the same price.
Multilingual E5 Large turns text in more than 90 languages into a fingerprint of its meaning, so a question in one language can find a matching passage in another. Use it for search across multilingual libraries and for comparing texts in different languages, with each input up to about half a page. Start questions with "query: " and stored passages with "passage: ". For English-only collections, E5 Large v2 is the English-only model with the same level of detail.
Deposit dinars with your Edahabia or CIB card. Every model, Multilingual-E5-Large included, is paid from that balance. No foreign card, no conversion.
The model card tags 94 languages, Arabic and French among them. Darja is not named.
Up to 512 tokens. Split anything longer.
1 024
Intfloat. Released 18 November 2025.
E5 Large v2 turns English text, and only English, from single sentences to whole documents, into a fingerprint of its meaning, so search finds text that means the same; it also suits reordering search results. Each input can run to about half a page. Start each search question with "query: " and each stored document with "passage: ", which is how the model was trained.
E5 Base v2 turns English text, and only English, into a fingerprint of its meaning, so search finds text that means the same; it also suits comparing texts and grouping similar ones. Each input can run to about half a page. Put "query: " before search questions and "passage: " before the documents you store, as the model was trained that way.
From what its maker and two benchmark sites publish, and where it stands on OpenDunes. Read 7 Nov 2026.
Open weights
The model's weights are public.
Index your help centre for semantic search
Find duplicate support tickets
Match product titles across two catalogues