by MetaReleased 12 October 2026
#1 for codeMuse Wonder is Meta's model for long, complex automated work, and it thinks before it answers. It takes PDFs, audio, video, pictures and text, answers in text, and keeps about a thousand pages in mind. It plans, hands tasks to helper agents, uses several tools at once, recovers from errors, operates browsers and desktop or mobile apps, and edits across many files in a code project.
You pay for what each request uses. No subscription, no minimum.
One message, the ordinary use
Every rate
| Meter | ↑ you send | ↓ comes back |
|---|---|---|
| Text tokens | 1 331,34 DA | 6 656,71 DA |
| Cached tokens read | 66,57 DA |
per 1M tokens
Web search · 0,83 DA per call
One company runs this model. OpenDunes sends every request to it. DA per 1M tokens: ↑ what you send · ↓ what comes back.
| Provider | ↑ Sent | ↓ Back | Cache read | Latency | Throughput | Uptime |
|---|---|---|---|---|---|---|
| Meta | 1 331,34 DA | 6 656,71 DA | 66,57 DA | … | … | … |
Also charged: a web search 0,83 DA each.
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 LabTurn this customer email into JSON with intent, urgency and order_id.
customer-email.txtTry 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="meta/muse-wonder",
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_tokenstemperaturetop_ptop_krepetition_penaltyReasoning
Let it think before answering — and see the thinking.
reasoningreasoning_effortinclude_reasoningStructured output
Get JSON you can parse instead of prose.
response_formatstructured_outputsTools
Let the model call your functions.
toolstool_choiceAbout 4,66 DA per message. ↑ 1 331,34 DA · ↓ 6 656,71 DA per 1M tokens.
No. You pay per use from your balance, with no subscription. Try it in Chat at the same price.
#1 for code · Muse Wonder is Meta's model for long, complex automated work, and it thinks before it answers. It takes PDFs, audio, video, pictures and text, answers in text, and keeps about a thousand pages in mind. It plans, hands tasks to helper agents, uses several tools at once, recovers from errors, operates browsers and desktop or mobile apps, and edits across many files in a code project.
Deposit dinars with your Edahabia or CIB card. Every model, Muse Wonder included, is paid from that balance. No foreign card, no conversion.
Meta names no languages for Muse Wonder. Arabic, French and Darja: not stated.
It reads text, images, video, audio and PDFs. Up to 1M tokens in one request, about 1 000 pages. It writes up to 708 000 words at once.
No. Meta has not published its weights.
Meta. Released 12 October 2026.
Muse Image makes and edits pictures from a written description, with optional reference pictures. It thinks before it draws, breaks multi-part requests into steps, and can search the web for factual requests. It writes readable text inside the picture, keeps a style or subject steady across a series, and edits by taking the previous result back with a fresh instruction.
Muse Voice Transcribe 1.0 turns speech into text in a single step. Send a single-channel WAV recording of up to 10 minutes and get the transcript back. It handles push-to-talk, finding where a speaker stops, and separate speakers, and you can steer it toward specialist terms or a language with a hint. It gives no word-by-word timing or confidence scores, so convert recordings in other formats to WAV first.
Muse Spark 1.3 Contributor is the lower-cost tier of Muse Spark 1.3, with the same text, picture, video and PDF input and thousand-page memory, for experiments, learning and early automated and coding projects. It keeps track of information across long tasks and works through conflicting information. For the previous release at this tier, use 1.2 Contributor.
#2 for websites · Muse Spark 1.3 is a Meta model that thinks before it answers, reading text, pictures, video and PDFs and replying in text, with about a thousand pages in mind at once. It is built for long-running automated and coding work: it keeps track of facts across a long task, works through conflicting information and asks you to clarify or confirm when it needs to, keeping its steps concise. Meta does not use your requests to improve its products.
From what its maker and two benchmark sites publish, and where it stands on OpenDunes. Read 7 Nov 2026.
Thinks first
Works through the problem before it answers.
Uses your tools
Can call functions your code provides.
Fills your JSON
Answers in the JSON shape you give it.
Searches the web
Can look things up online while it answers.
Cheaper on repeats
A prompt it saw a moment ago costs less the second time.
Plan a multi-file refactor, delegate the steps and recover from failed tests
Thinks first · Uses your tools
Build a polished website or interface component from a written brief
Search the web as it reasons and cite what it found
Searches the web