by ScaledownReleased 13 October 2026
ScaleDown Summarize shortens long text and documents by rewriting them faithfully in new words, not by copying sentences out. It reads about 1,000 pages at once and writes back about 4 pages at most. Use it for reviewing documents, digesting transcripts, or trimming retrieved material before another model sees it, steering layout, focus, language and length with instructions.
You pay for what each request uses. No subscription, no minimum.
20 % offScaleDown has cut its price. Every figure here is the new one.
One message, the ordinary use
Every rate
| Meter | ↑ you send | ↓ comes back |
|---|---|---|
| Text tokens | 13,31 DA |
per 1M tokens
One company runs this model. OpenDunes sends every request to it. DA per 1M tokens: ↑ what you send.
| Provider | ↑ Sent | Latency | Throughput | Uptime |
|---|---|---|---|---|
| ScaleDown20 % off | … | … | … |
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.
Real requests, real replies. Try any of them yourself — one click, pre-filled.
What's the total on this invoice, and when is it due?
invoice-scan.jpg · photoThe total is 42 300,00 DA, due 5 September 2026. Worth a double-check: the handwritten day could read as 8 instead of 3.
Try in Chat LabTwo lines to welcome a shop in Oran.
Welcome to our new shop in Oran — fresh bread every morning, and coffee the way the city likes it. Come in, the first cup is on us.
Try in Chat LabTry your own request — same model, pre-selected.
Open Chat LabSwap 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.chat.completions.create(
model="scaledown/summarize",
messages=[{"role": "user", "content": "Two lines to welcome a shop in Oran."}],
)
print(resp.choices[0].message.content)Runs this request once on your balance.
Standard OpenAI fields, grouped by what they control. Anything not listed is accepted and ignored.
Shaping the reply
How long, how varied, where to stop.
max_tokensAbout 0,013 DA per message. Was 0,016 DA. ↑ 13,31 DA per 1M tokens.
No. You pay per use from your balance, with no subscription. Try it in Chat at the same price.
ScaleDown Summarize shortens long text and documents by rewriting them faithfully in new words, not by copying sentences out. It reads about 1,000 pages at once and writes back about 4 pages at most. Use it for reviewing documents, digesting transcripts, or trimming retrieved material before another model sees it, steering layout, focus, language and length with instructions.
Deposit dinars with your Edahabia or CIB card. Every model, Summarize included, is paid from that balance. No foreign card, no conversion.
Scaledown names no languages for Summarize. Arabic, French and Darja: not stated.
It reads text, images and PDFs. Up to 1M tokens in one request, about 1 000 pages. It writes up to 3 000 words at once.
No. Scaledown has not published its weights.
Scaledown. Released 13 October 2026.
Classify-1 reads text, pictures or whole documents and answers questions you set: pick one option from several, yes or no, or rate on an ordered scale. It returns a probability for each answer rather than written text, which suits sorting requests, moderation, spotting what a customer wants and tagging by topic. It reads about 1,000 pages at once. To pull specific values out of a document use ScaleDown Extract, and to shorten it use ScaleDown Summarize.
ScaleDown Extract pulls the details you name out of text, pictures and PDFs. You list the fields you want in plain English, and it returns each match with a confidence score, its position in the document and the text around it. It suits contracts, invoices and emails, plus transcripts, and it reads scanned pages itself.
From what its maker and two benchmark sites publish, and where it stands on OpenDunes. Read 7 Nov 2026.
Thinks first No.
Works through the problem before it answers.
Uses your tools No.
Can call functions your code provides.
Fills your JSON No.
Answers in the JSON shape you give it.
Searches the web No.
Can look things up online while it answers.
Cheaper on repeats No.
A prompt it saw a moment ago costs less the second time.
Condense a meeting transcript into action items for the team
Shorten retrieved passages before they reach a larger model
Summarize a legal brief inside a word limit and a language you set