Meta

Muse Wonder

Models.dev
Intelligence
43.6
Rank 45 of 790 measured models
Output speed
110 t/s
Rank 147 of 456 measured models
Price/1M
$8
Rank 673 of 741 measured models
Context window
1.05M
Rank 50 of 1368 measured models
Cost per task
$6.95
Rank 222 of 235 measured models

Muse Wonder (Max) scores 43.6 on the Artificial Analysis Intelligence Index, ranking 45th of 790 measured models. The closest models on this index are Claude Opus 5 (44.2), GPT-6 Sol (44.2), and Claude Opus 4.8 (batch) (41.8). At $8 per million tokens (3:1 input/output blended price), it is cheaper than Claude Opus 5 ($10), more expensive than GPT-6 Sol ($4) and Claude Opus 4.8 (batch) ($5). With an output speed of 110 t/s, it is faster than Claude Opus 5 (81.2 t/s), about as fast as GPT-6 Sol (111 t/s).

Artificial Analysis Indices
Maximum
Intelligence43.6
Rank 45 of 790 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Cost per task
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting51.1
Rank 33 of 260 measured models
Strategy & Ops54.5
Rank 30 of 260 measured models
Legal46.9
Rank 53 of 260 measured models
Healthcare & Medical48.7
Rank 21 of 190 measured models
Engineering49.6
Rank 39 of 259 measured models
Economics49.5
Rank 50 of 267 measured models
Standard Benchmarks
FrontierCode 2.027.9%
Rank 15 of 33 measured models
Terminal-Bench-Science25.7%
Rank 12 of 33 measured models
HLE-Diamond39.0%
Rank 11 of 33 measured models
CritPt v2 (Hard Subset)50.5%
Rank 11 of 29 measured models
AutomationBench-AA68.1%
Rank 10 of 33 measured models
GDP.pdf (all-pass)30.6%
Rank 5 of 33 measured models
HLE50.1%
Rank 26 of 719 measured models
SciCode60.3%
Rank 22 of 286 measured models
Additional benchmarks12
Design Arena · 3d60.7%
Design Arena · Blender3d36.0%
Design Arena · Codecategories62.5%
Design Arena · Codetovideo56.0%
Design Arena · Dataviz55.1%
Design Arena · Gamedev62.1%
Design Arena · Htmlslides14.0%
Design Arena · Pdf50.7%
Design Arena · Python Pptxslides13.2%
Design Arena · Spreadsheet46.2%
Design Arena · Uicomponent61.3%
Design Arena · Website63.8%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy