Muse Spark vs Mistral Large 4.0: Pratinjau versus Model yang Telah Dirilis

This is not a fair fight, and M​istral would be the first to say so. M​istral Large 4.0 was announced on October 6, 2026 as a public preview under the nickname "le Chonk" — a 1-trillion-parameter mixture of experts with 52 billion active parameters, self-described as still improving, with the weights promised by the end of October. M​use Spark 1.3, M​eta's current M​use Spark checkpoint, shipped on September 2, 2026 and has been stable since. Comparing a preview model against a shipped one is a category error if you are picking production infrastructure and a genuinely useful exercise if you are trying to work out which of the two open-weight-adjacent lines deserves your attention over the next two quarters. Which of those you are doing changes every conclusion below, so this piece keeps the two questions separate.

Di mana sebenarnya posisi masing-masing dari mereka

M​use Spark is a family, not a model. Muse Spark 1.1 arrived July 16, 2026, Muse Spark 1.2 on August 5, and M​use Spark 1.3 on September 2, all from M​eta Superintelligence Labs, all sharing a 1,048,576-token context window. M​use Spark 1.3 gained its max reasoning setting on September 4. It is live, priced, documented and stable — and the family's pricing has not moved in three consecutive releases.

M​istral Large 4 is at the other end of its lifecycle. The announcement is unusually candid about it: a public preview on M​istral Studio today, weights "by the end of the month", a model that "continues to improve rapidly as we refine it", currently being red-teamed with cybersecurity partners and vetted organisations under reduced moderation. It was trained from scratch on 3,800 N​VIDIA Grace Blackwell GPUs in M​istral's own European datacentres, and M​istral is explicit that sovereignty is the point — a European-hosted, self-deployable model for organisations that cannot accept provider-level refusals mid-incident.

Perbedaan dalam kematangan itu adalah hal pertama yang harus diperhitungkan, dan itu bukanlah sebuah celaan. Pratinjau yang akan berubah adalah pratinjau yang belum seharusnya Anda jadikan dasar untuk membangun pipeline tersertifikasi; model yang sudah dirilis yang tidak berubah selama enam minggu adalah model yang bisa Anda jadikan dasar.

Angka-angka independen

Both rows are Artificial Analysis' hosted measurements; neither is a vendor claim, and the listing names M​istral's entry "M​istral Large 4 Preview".

• Intelligence Index — M​use Spark 1.3 (max): 48, rank 24 of 217. M​istral Large 4 Preview: 38, rank 66 of 217.

• Cost per index task — M​use Spark 1.3: $1.60. M​istral Large 4 Preview: $1.13. M​istral is cheaper per completed task despite the lower score.

• Output speed — M​use Spark 1.3: 121.9 tokens per second, rank 43 of 186. M​istral Large 4 Preview: not measured, no figure published. On a preview served from a single in-house cluster, an unmeasured speed is unsurprising and still a gap.

• Verbosity — M​use Spark 1.3: 170M output tokens on the index. M​istral Large 4 Preview: 200M, described as very verbose. Advantage M​use Spark.

• List pricing — M​use Spark 1.3 at $1.25 in / $4.25 out with cached input at $0.15, flat. M​istral Large 4 is currently on a sale rate reported at $0.68 in / $2.09 out with cached input at $0.07, against regular rates of $1.36 and $4.18. M​istral's own public pricing page still shows a generic "M​istral Large" figure and does not list Large 4, so the documentation page is the authoritative source and the sale rate is the one to verify before you commit.

Baca tabel itu sebagai pernyataan tentang hari ini dan tabel itu mengatakan sesuatu yang agak janggal bagi M​eta: model yang lebih murah di sini adalah model yang skornya lebih rendah, dan skornya lebih rendah sepuluh poin indeks. Baca tabel itu sebagai pernyataan tentang pratinjau dan tabel itu mengatakan sesuatu yang lebih berguna — biaya operasionalnya kompetitif sementara kemampuannya belum memadai, yang merupakan gambaran model yang masih punya ruang untuk berkembang dan vendor yang menetapkan harga selaras dengan itu.

Konflik jendela konteks, dinyatakan alih-alih diselesaikan

M​istral's own model card gives M​istral Large 4 a 1-million-token context window. Artificial Analysis' listing for the same model shows 524,000 in its specification panel and describes it as "520k tokens" in the prose. We are not going to average those. One of the two is wrong, or one is describing a configuration the other is not, and we could not resolve it from the public record. What we can say is that the vendor figure is more specific about modality and architecture and the analyst figure is more specific about cost and task performance, so neither source dominates. If window size matters to your decision, verify it against the API before you commit — this is exactly the kind of discrepancy that propagates into production unnoticed.

On modalities there is no conflict and no ambiguity: both models take text and image in and return text. M​istral Large 4's model card adds reasoning as an output modality and lists structured outputs, function calling, document Q&A, prefix completion, batching, agents and connectors among its features. M​use Spark 1.3 takes text, image and video in and returns text, with a reasoning ladder from minimal to max. Notably, neither model publishes a maximum output token figure that we could verify — which is one of the few places in the current field where a frontier model from a major lab simply does not say.

Apa yang diklaim masing-masing vendor, tetap berada di sisinya sendiri.

M​istral's announcement leans on vertical strengths rather than a general scoreboard. Vendor-reported and unreproduced: DeepSWE v1.1 at 61.7%, SWE-Atlas-QnA at 59.4%, Terminal-Bench 4 at 28.3%, a Combined Coding Agent Index of 49.8%, AutomationBench at 59.9%, AA-Briefcase at 1,393 Elo, a second-place 3.74 out of 5 in Surge AI's blind human coding evaluation, 42% on Dense 200 visual grounding, 93.3% attack resistance on Lakera B3, and a cybersecurity pitch that is the real differentiator — top five on the Artificial Analysis Cyber Index, 82% on a reproduce-and-patch task that M​istral claims is the highest of any model, and 93% on Cybench. M​istral's own framing of that last point is that several leading closed models, Claude Opus 5.5 and GPT-6 Astra among them, score near zero on the same test because they refuse the task rather than fail it.

M​eta's figures for M​use Spark 1.3 are a general-purpose agentic scoreboard: DeepSWE v1.1 at 75.4, OSWorld 2.0 at 66.9, Terminal-Bench 2.1 at 76.4 with the Terminus 2 agent, MCP-Atlas at 89.2, GDPval-AA v2 at 1,754, SciCode at 58.5, HLE-Diamond without tools at 26.1 and JobBench at 61.8.

Kedua daftar itu nyaris tidak bersinggungan. DeepSWE v1.1 adalah satu-satunya evaluasi yang diterbitkan keduanya, dan di sana Meta mengklaim 75,4 berbanding 61,7 milik Mistral — selisih yang cukup lebar untuk bertahan dari perbedaan harness, meskipun kami belum mereproduksi salah satu pun. Di tempat lain, para vendor hanya mengukur hal-hal yang berbeda, itulah sebabnya perbandingan yang jujur bertumpu pada baris-baris independen di atas, bukan pada menumpuk kedua tabel ini.

Pertanyaan yang sebenarnya membedakan mereka: bobot terbuka

Inilah sumbu yang disiratkan oleh judul pertandingan tersebut dan sumbu di mana kedua model itu benar-benar merupakan produk yang berbeda, bukan versi yang lebih baik dan lebih buruk dari hal yang sama.

M​istral Large 4 memiliki jalur yang dinyatakan menuju bobot yang dapat diunduh, dijanjikan pada akhir Oktober 2026, dengan klaim yang menyertainya bahwa model ini mengungguli model open-weight mana pun yang dikembangkan di AS atau Eropa. Jika itu terwujud, ini menjadi rilis open-weight Eropa terbesar dan opsi nyata bagi organisasi yang perlu melakukan self-host. Perlu dicatat bahwa lisensi tidak disebutkan namanya di kartu model pada saat tulisan ini dibuat dan repositori Hugging Face tidak dapat dibaca publik saat kami memeriksanya, jadi ketentuan yang menyertai kehadiran bobot tersebut belum dapat diketahui — celah yang tidak akan kami tutupi, karena dalam rilis open-weight, lisensi sering kali justru menjadi inti persoalannya.

The M​use Spark line is not an open-weight line. Muse Spark 1.1, 1.2 and 1.3 are closed and served through M​eta's own API. What does carry weights is a separate member of the family under its own name — M​use Spark Open, an 816B-total / 42B-active release whose repository was created on September 15, 2026, sat gated and unannounced for four weeks, and was published with a licence by M​eta on October 15, 2026. That timeline is the part we can state flatly, because the repository timestamps and the licence date are in the artefact. What is not confirmed is how good it is: M​eta has not published the evaluation report behind its own scorecard, so treat the Open weights as their own question rather than as a newer version of M​use Spark 1.3. Anyone comparing M​istral's open-weight promise against M​use Spark 1.3 is comparing two different distribution models, not two capability tiers.

Which means the decision rule is simple and has almost nothing to do with the index. If you must run the model on your own hardware, in your own jurisdiction, under your own policies, this comparison has one candidate and it is the M​istral one, subject to what its licence says when the weights appear. If you are buying an endpoint, the M​istral preview is currently the lower-scoring and cheaper option against a shipped model that beats it by ten index points.

Membangun terhadap salah satu tanpa mengunci pilihan

OrcaRouter routes Muse Spark 1.2 and Muse Spark 1.1 at Meta's list price with 0% markup passed through, alongside 200+ models on one API, with automatic failover and a routing DSL for splitting traffic between them. M​istral Large 4 is not on our catalogue, and M​use Spark 1.3 is not either — so we are not going to pretend this matchup can be run end-to-end on one of our keys today. What the pass-through model is good for here is the specific thing this matchup needs: as M​istral's sale pricing and its weight release both land within weeks of each other, a routed key means a price change or a new open build shows up on our side the same day it shows up on theirs, rather than at the next contract renewal.

Apa yang harus dilakukan bulan ini

If your decision can wait until the end of October, wait — the single most consequential fact about M​istral Large 4 is not yet published, and it is the licence on the weights rather than any number in this article. If it cannot wait, the comparison as it stands today is between a preview at $0.68 / $2.09 on sale with a 38 index score and no measured output speed, and a shipped model at $1.25 / $4.25 with a 48 index score, roughly 122 tokens per second and six weeks of stability behind it. The preview is cheaper per task today and will almost certainly be a different model by the time anyone reads this; the shipped one is ten points stronger and will not have changed. Neither of those is a recommendation, because the deciding factor is whether the weights M​istral promised are something you need — and that is not a benchmark question.

Hasil seperti ini juga menjadi pengingat bahwa kesalahan model sudah terdokumentasi: insiden AI dan kegagalan agen di balik model di kelas ini dikumpulkan dalam the OrcaRouter AI Incident Archive, dengan setiap entri mencantumkan sumbernya sendiri, tingkat keparahan, peringkat keyakinan, dan penanda yang disengketakan.