Akses API Muse Wonder dan Harga: Apa yang Didapat dengan $4/$20 pada Model Frontier Pertama Meta

Meta Superintelligence Labs mulai menjual Muse Wonder pada 12 Oktober 2026, dengan harga $4,00 per juta token masukan dan $20,00 per juta token keluaran — dan bagian rilis yang lebih canggih, Muse Wonder Full, sama sekali tidak memiliki harga per token. Itulah keseluruhan kisah aksesnya dalam satu kalimat: dua checkpoint dikirim, dua pintu berbeda, dan hanya satu di antaranya yang terbuka dengan kunci API.

So this page answers the questions that come before any benchmark comparison. Which identifier you can call, what the rate card contains beyond the two headline numbers, what M​eta says the model does, what an outside evaluator has measured so far, and what the current answer is if you wanted to reach it from a router today. Nothing on this page implies OrcaRouter serves Muse Wonder — it does not, and we say so in the section where it matters.

Muse Wonder mana yang sebenarnya bisa kamu telepon?

M​eta's material describes a series rather than a model. The two members behave differently as products, and the difference is administrative rather than technical.

• muse-wonder-1.0 — the checkpoint served on the Meta Model API and inside Muse Code, M​eta's terminal coding agent. This is the identifier a developer provisions against, and every per-token figure below belongs to it.

• muse-wonder-full-1.0 — a separate checkpoint that M​eta describes as the higher-scoring configuration. It is not an effort flag on the API model and it is not a second API endpoint; M​eta sells it through a Muse Code subscription. M​eta's published plans for that product are $50 a month for Power Usage and $200 a month for Maximum Usage, which means Full's real unit of purchase is a seat-month, not a million tokens.

Itu adalah bentuk yang tidak biasa untuk sebuah rilis frontier, dan ada konsekuensi praktis yang layak dinyatakan secara terus terang: tidak ada jalur API menuju checkpoint dengan skor yang lebih baik. Tim yang melakukan benchmark pada model API dan menyukainya tidak bisa sekadar memutar tombol dan mendapatkan Full. Mereka harus membeli langganan, atau menerima bahwa model yang mereka ukur adalah model yang mereka dapatkan.

Kartu tarif, baris demi baris

M​eta's per-token pricing applies to the API checkpoint. The rows below are vendor rate-card facts unless marked otherwise, all per million tokens.

• Input — $4,00 per juta token

• Input yang di-cache — $0,20 per juta token, yang oleh evaluator independen Artificial Analysis dibaca sebagai diskon cache 95%

• Keluaran — $20,00 per juta token

• Biaya gabungan pada campuran cache-hit/input/output 7:2:1 — sekitar $2,94 per juta token, menurut halaman model Artificial Analysis sendiri

• Measured cost of completing one Intelligence Index task — about $4.52, again Artificial Analysis' figure rather than M​eta's

• Jendela konteks — 1.048.576 token dinyatakan dan dilayani

• Output maksimum dalam satu respons — 943.718 token, yang merupakan batas atas respons tunggal terbesar di papan yang kami pantau

• Modalitas masukan — teks, gambar, ucapan, dan video, dibandingkan dengan keluaran teks

• Reasoning control — mandatory. M​eta exposes an effort ladder from minimal to max, defaulting to medium, but not an off switch

Satu detail struktural mudah terlewat dan penting bagi siapa pun yang menghitung biaya retrieval atau pekerjaan agen panjang: tarif input tidak berjenjang pada ambang token. Prompt 400.000 token ditagih dengan tarif per token yang sama seperti prompt 400 token. Beberapa model unggulan pesaing menetapkan harga ulang untuk seluruh permintaan begitu prompt melewati batas yang didokumentasikan, jadi ini benar-benar perbedaan pada bentuk kurvanya, bukan hanya pada tingginya.

What M​eta says it scores, and what that is worth

M​eta shipped a benchmark table covering both checkpoints at their maximum reasoning settings, and it is the only scoreboard that covers Muse Wonder Full at all. It is vendor-reported, it was produced in M​eta's own harness, and none of it has been independently reproduced — read the following as M​eta's claim rather than as a measurement.

The rows M​eta promotes are the sustained-agent ones. On SWE-Marathon, a long-horizon software task set, M​eta places Muse Wonder Full ahead of every model in its comparison set. On DeepSWE 1.1 it makes a similar claim, and on the composite index it publishes the series carries its own name. Independent commentary has repeated those two rows selectively, and at least one widely shared post paired a Muse Wonder Full figure with a lower competitor number without noting that the underlying table is the vendor's, unreproduced, and unreleased as a harness.

The awkward part is internal to the table. Muse Wonder Full is not uniformly better than the cheaper API checkpoint: the two profiles trade rows, which suggests M​eta trained toward two different objectives rather than turning one dial up. If that is right, then "Full is the better model" is not a safe assumption for your own workload, and the only way to know which profile fits is to measure both — which currently means buying the subscription.

Apa yang telah diukur oleh evaluator eksternal

Artificial Analysis is the only evaluator with a published row for Muse Wonder as of October 13, 2026, and its numbers are a different species from M​eta's: same harness for every model on the board, published configuration, and no vendor input.

• Intelligence Index — 52 pada upaya maksimum, peringkat kesembilan di papan yang mencakup lebih dari 200 model

• Biaya per tugas indeks — sekitar $4.52, yang ditandai evaluator sebagai di atas median untuk kohort setelah harga diperhitungkan

• Token output yang dikonsumsi pada eksekusi tersebut — sekitar 130 juta dibandingkan median papan sekitar 85 juta, yang menurut evaluator agak bertele-tele

• Kecepatan penyajian — sekitar 115 token keluaran per detik, dengan waktu ke token pertama mendekati 20 detik

• Muse Wonder Full — indeks 56 dengan upaya maksimal, diperingkat dekat puncak papan, tanpa harga per token karena tidak dijual per token

Note the last row against M​eta's table: on the composite index M​eta publishes 59.0 for Muse Wonder Full while the independent board reads 56. Three points on a scale like that is not a rounding difference, and the independent figure is the one to plan against, because it was produced the same way as the numbers for every other model you are weighing.

Di mana Anda belum dapat meneleponnya

Muse Wonder is available through M​eta's own API and inside Muse Code, and that is currently the whole list. There is no third-party serving coverage in the catalogue we can verify, and it is not in OrcaRouter's catalogue — a router cannot route a checkpoint nobody else is serving, and we would rather say that than imply otherwise.

What that means in practice is a two-vendor setup if you want to try it: M​eta's key for the new model, and whatever you already use for everything else. That is the situation a router is built for on the other side of the fence. OrcaRouter passes provider list price through with 0% markup across more than 200 models behind one key, which includes M​eta's own Muse Spark 1.2 and Muse Spark 1.1 — so the M​eta models you can call today are already reachable in one place, and a rate change at M​eta lands here the same day it is announced rather than waiting for a middleman to re-price. When a second provider does start serving Muse Wonder, automatic failover across endpoints is what keeps a single unhealthy path from taking your traffic down.

Apa yang akan mengubah jawaban ini

Ada empat hal yang dapat diperiksa, dan salah satunya saja akan membuat halaman ini layak dibaca ulang daripada menebak-nebak lagi.

• Pengenal model yang muncul untuk Muse Wonder Full di Meta Model API, yang akan mengakhiri batas khusus langganan

• Revisi terhadap indeks independen yang mengubah 52, atau kedatangan evaluator kedua dengan barisnya sendiri

• Perubahan harga pada salah satu baris kartu tarif, yang pada markup 0% aktif di sisi kami pada hari yang sama

• Muse Wonder muncul di katalog pihak ketiga dengan tarif yang dipublikasikan, yang akan menjadikan pertanyaan ketersediaan sebagai pertanyaan perutean, bukan pertanyaan vendor

Until then the honest summary is that M​eta has priced its API checkpoint 20% under the current A​nthropic and O​penAI flagships on input and output alike, made its cheaper model the one you can reach programmatically, and left the interesting configuration behind a monthly seat. The $4/$20 is real and it is the cheapest frontier rate M​eta has ever published. It is also the price of the model you can actually call.