opendunes

Solar Pro 4

by UpstageReleased 10 August 2026

Upstage's Solar Pro 4, a large model with a 524K context for long-horizon tasks and agent workflows, strong in office productivity, document-heavy work and coding, above Solar Mini 4

TakesText
GivesText
Open in chatCreate API keyCompare

What it costs

You pay for what each request uses. No subscription, no minimum.

≈ 0,15 DA per message3 000 DA buys about 20 000 messages

50 % offUpstage has cut its price. Every figure here is the new one.

One message, the ordinary use

1 000 tokens sent× 49,93 DA per 1M0,050 DA
500 tokens back× 199,70 DA per 1M0,10 DA
Total≈ 0,15 DA

Every rate

Meter↑ you send↓ comes back
Text tokens49,93 DA199,70 DA
Cached tokens read9,99 DA

per 1M tokens

Answer a quick question≈ 0,057 DA
Reply to a customer email≈ 0,11 DA
Translate a full page of text≈ 0,19 DA
Review a pull request of about 400 lines≈ 0,54 DA
Summarise a 20-page report≈ 0,84 DA
When you pay
After the answer, from your balance. Nothing is held.
Why it varies
A longer question or a longer answer costs more.
After each one
The exact charge shows on the request.
DepositDeposits from 50 DA, with Edahabia or CIB.

Where your request runs

One company runs this model. OpenDunes sends every request to it. DA per 1M tokens: ↑ what you send · ↓ what comes back.

Upstage · Zdr50 % off…
↑ Sent, per 1M
Was 99,8649,93 DA
↓ Back, per 1M
Was 399,40199,70 DA
Cache read
Was 19,989,99 DA
Latency
…
Throughput
…
Upstage50 % off…
↑ Sent, per 1M
Was 99,8649,93 DA
↓ Back, per 1M
Was 399,40199,70 DA
Cache read
Was 19,989,99 DA
Latency
…
Throughput
…
Provider↑ Sent↓ BackCache readLatencyThroughputUptime
Upstage · Zdr50 % offWas 99,8649,93 DAWas 399,40199,70 DAWas 19,989,99 DA………
Upstage50 % offWas 99,8649,93 DAWas 399,40199,70 DAWas 19,989,99 DA………

You pay the price of the provider that served your request. Prices read 15:48 · speed over the last 30 minutes · uptime over 3 days.

See it in action

Real requests, real replies. Try any of them yourself — one click, pre-filled.

Structured output
You send

Turn this customer email into JSON with intent, urgency and order_id.

customer-email.txt
It replied

{ "intent": "refund", "urgency": "high", "order_id": "OD-58421" }

Try in Chat Lab
Writing
You send

Two lines to welcome a shop in Oran.

It replied

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 Lab

Try your own request — same model, pre-selected.

Open Chat Lab

Call it in code

Swap the base URL and the key. Your OpenAI code runs unchanged.

POST /v1/chat/completions
Base URL
https://opendunes.com/v1
Model ID
upstage/solar-pro4
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="upstage/solar-pro4",
    messages=[{"role": "user", "content": "Two lines to welcome a shop in Oran."}],
)

print(resp.choices[0].message.content)
pip install openaiFull reference for this route
Sign in to run it

Runs this request once on your balance.

Parameters you can send

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_tokenstemperaturetop_pfrequency_penaltypresence_penalty

Reasoning

Let it think before answering — and see the thinking.

reasoningreasoning_effortinclude_reasoning

Structured output

Get JSON you can parse instead of prose.

response_formatstructured_outputs

Tools

Let the model call your functions.

toolstool_choiceparallel_tool_calls

Questions about Solar Pro 4

  • How much does Solar Pro 4 cost in dinars?

    About 0,15 DA per message. Was 0,30 DA. ↑ 49,93 DA · ↓ 199,70 DA per 1M tokens.

  • Is Solar Pro 4 free? Can I try it?

    No. You pay per use from your balance, with no subscription. Try it in Chat at the same price.

  • What is Solar Pro 4 best for?

    Upstage's Solar Pro 4, a large model with a 524K context for long-horizon tasks and agent workflows, strong in office productivity, document-heavy work and coding, above Solar Mini 4.

  • How do I pay for Solar Pro 4 without an international card?

    Deposit dinars with your Edahabia or CIB card. Every model, Solar Pro 4 included, is paid from that balance. No foreign card, no conversion.

  • Does Solar Pro 4 work in Arabic, French or Darja?

    Upstage names no languages for Solar Pro 4. Arabic, French and Darja: not stated.

  • Can it read PDFs, images or a whole book?

    It reads text. Up to 524K tokens in one request, about 520 pages. It writes up to 98 000 words at once.

  • Is Solar Pro 4 open source?

    No. Upstage has not published its weights.

  • Who makes Solar Pro 4, and when was it released?

    Upstage. Released 10 August 2026.

Other Upstage models, priced in dinars

See all Upstage models
Solar DecideUpstage · Sep 28, 2026

Solar Decide does not write text. You describe a situation, ask typed questions, and it answers with one pick from your options, a number, or a yes or no, each with a probability. A whole document of about 500 pages can serve as the situation, and the Korean it understands is strong, which helps in routing, sorting and policy checks. Solar Decide Flash is the faster variant.

524 pages per call
≈0,017 DAper case
View model
Solar Decide FlashUpstage · Oct 8, 2026

Solar Decide Flash does not write text. You describe a situation, ask typed questions, and it answers with one pick from your options, a number, or a yes or no, each with a probability. It is tuned for fast, even response times in routing, sorting and policy checks, and a document of about 500 pages can serve as the situation. Solar Decide is the standard version when speed matters less.

524 pages per call
≈0,017 DAper case
View model
Solar Mini 4Upstage · Sep 23, 2026

Solar Mini 4 reads text and writes text, keeping about 520 pages in mind at once. It is Upstage's compact, low-cost model for jobs where quick answers matter, such as work using the tools you connect to it. It is fluent in Korean, strong in English and Japanese, and copes with long material.

Thinks firstUses your tools520 pages at once
≈0,050 DAper message
View model