Muse Wonder

Meta, closed weights, released Oct 2026, run at provider defaults through OpenRouter, served by Meta. It built the same five pages as every model on season 0.5 of the Rams Index, and Rams graded each one at desktop and phone size.

Rams Index, 0 to 100

60.5

95% interval 53.2–68.9

Rank: 7of 14 models by Index; the paired intervals allow 1 to 12
Cost per page: $0.25Generation only, list price on run day
Second judge: 74.8GPT-5.6 Terra places it eighth of 14. The Claude judge: seventh.
Grade sum: 23.95Seven 1–5 grades, out of 35
Pages judged: 1010 judge passes, 2 builds per brief
Defects per page: 8.5Low-contrast text (10 of 10)

Rank 7 of 14; the data allows anywhere from 1 to 12. Within noise of GPT-6 Astra, Grok 4.7, GPT-6.1 Sol, Claude Sonnet 5.5, GPT-6 Luna, Gemini 4 Argon, Claude Opus 5.5, GPT-5.6 Sol, Claude Fable 5.1, Claude Fable 5 and DeepSeek V4 Pro: the gap to each sits inside the noise, so the board can't order them yet. Not clearly behind any model. Clearly ahead of Gemini 3.8 Flash and Claude Opus 4.8.

Designer check pending · read Nov 6Season 0.5New this seasonProvisional: 1.0 judge passes per page, protocol 2

Grades by dimension

The seven grades behind the Index, 1 to 5, mean of desktop and phone. The tick is the mean across all 14 models.

Hierarchy3.508 of 14
Typography3.357 of 14
Spacing3.157 of 14
Color3.554 of 14
Components3.507 of 14
Composure3.506 of 14
Craft3.406 of 14

Bars start at 2.8 and end at 4.2, the range the board uses, so small differences show.

Index by brief

The same Index on each of the five briefs, and where the model places on that brief among all 14.

Landing63.49 of 14
Dashboard61.64 of 14
Settings54.52 of 14
Pricing49.113 of 14
Article74.17 of 14

Bars run from 40 to 90.

How it ran

Every model runs at its provider's default settings, through the same harness, with the same budget per page.

Effort: DefaultThe provider’s default. We send no effort or thinking setting
Time per page: 125sMedian over every build, broken ones included
Output tokens: 12.4kMedian per page, reasoning included
Reasoning tokens: 1.1kMedian per page, as the API reports them
Broken pages: 0 of 10Every build came back as a whole page
Criticals: 20%2 of 10 scored pages had a critical issue in their first Rams code review
Built through: OpenRouterServed by Meta

Best and worst page

Of its 10 judged pages, the one the judge graded highest and the one it graded lowest, with the judge's top findings on each, most severe first.

Best: Article82.1
Index on this page, one judge pass. Brief mean 74.1.
- Deck text weight competes with drop-cap paragraphModerate
- Tag row vertical rhythm slightly tight against metadata rowModerate
- Topbar center span text truncated in viewportModerate
Worst: Pricing44.6
Index on this page, one judge pass. Brief mean 49.1.
- Redundant 'no credit card' messageSerious
- Competing accent color events: purple used ubiquitouslySerious
- Redundant 'no credit card' messagingSerious

All 10 of its pages sit beside every other model's in the gallery.

Head to head

Muse Wonder against every other model: the paired difference in Index points, and whether the 95% interval on it clears zero.

vs GPT-6 Astra-5.9Inside the noise
vs Grok 4.7-5.5Inside the noise
vs GPT-6.1 Sol-4.8Inside the noise
vs Claude Sonnet 5.5-1.2Inside the noise
vs GPT-6 Luna-1.2Inside the noise
vs Gemini 4 Argon-1.1Inside the noise
vs Claude Opus 5.5+0.2Inside the noise
vs GPT-5.6 Sol+1.6Inside the noise
vs Claude Fable 5.1+1.6Inside the noise
vs Claude Fable 5+2.9Inside the noise
vs Gemini 3.8 Flash+5.0Clearly better
vs DeepSeek V4 Pro+5.9Inside the noise
vs Claude Opus 4.8+7.7Clearly better

The other models

Season 0.5 of the Rams Index, in board order, then the unranked.

Rams Index v0.5 · updated 2026-10-17Judged by Rams eyes · Claude Sonnet 4.6