Claude Fable 5

Anthropic, closed weights, released Jun 2026, run at provider defaults through the Anthropic API. 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

57.6

95% interval 51.0–67.0

Rank: 11of 14 models by Index; the paired intervals allow 5 to 13
Cost per page: $0.53Generation only, list price on run day
Second judge: 74.5GPT-5.6 Terra places it ninth of 14. The Claude judge: 11th.
Grade sum: 23.13Seven 1–5 grades, out of 35
Pages judged: 1520 judge passes, 3 builds per brief
Defects per page: 3.2Content wider than the phone (11 of 15)

Rank 11 of 14; the data allows anywhere from 5 to 13. Within noise of GPT-6 Luna, Gemini 4 Argon, Muse Wonder, Claude Opus 5.5, GPT-5.6 Sol, Claude Fable 5.1, Gemini 3.8 Flash and DeepSeek V4 Pro: the gap to each sits inside the noise, so the board can't order them yet. Clearly behind GPT-6 Astra, Grok 4.7, GPT-6.1 Sol and Claude Sonnet 5.5. Clearly ahead of Claude Opus 4.8.

Designer check pending · read Nov 6Season 0.5Provisional: 1.3 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.

Hierarchy3.3311 of 14
Typography3.2810 of 14
Spacing3.204 of 14
Color3.2712 of 14
Components3.3510 of 14
Composure3.3810 of 14
Craft3.3210 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.

Landing59.513 of 14
Dashboard56.09 of 14
Settings49.411 of 14
Pricing49.712 of 14
Article73.59 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: Default (always thinks)The provider’s default. We send no effort or thinking setting
Time per page: 99sMedian over every build, broken ones included
Output tokens: 10.3kMedian per page, reasoning included
Reasoning tokens: n/aIts API doesn’t report them separately
Broken pages: 0 of 15Every build came back as a whole page
Criticals: 20%3 of 15 scored pages had a critical issue in their first Rams code review
Built through: Anthropic APIServed by Anthropic

Best and worst page

Of its 15 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: Article83.9
Index on this page, one judge pass. Brief mean 73.5.
- Section heading register blends with body weightModerate
- Blockquote left-border decoration conflicts with column measureModerate
- Drop-cap 'E' sits uncomfortably close to body text column edgeModerate
Worst: Dashboard48.2
Index on this page, one judge pass. Brief mean 56.0.
- Bar chart uses generic library defaultsModerate
- Large dead zone below Recent Episodes tableModerate
- Stat card delta labels repeat 'vs prior 30d' four times identicallyModerate

All 15 of its pages sit beside every other model's in the gallery.

Head to head

Claude Fable 5 against every other model: the paired difference in Index points, and whether the 95% interval on it clears zero.

vs GPT-6 Astra-8.8Clearly behind
vs Grok 4.7-8.5Clearly behind
vs GPT-6.1 Sol-7.8Clearly behind
vs Claude Sonnet 5.5-4.2Clearly behind
vs GPT-6 Luna-4.2Inside the noise
vs Gemini 4 Argon-4.0Inside the noise
vs Muse Wonder-2.9Inside the noise
vs Claude Opus 5.5-2.8Inside the noise
vs GPT-5.6 Sol-1.3Inside the noise
vs Claude Fable 5.1-1.3Inside the noise
vs Gemini 3.8 Flash+2.1Inside the noise
vs DeepSeek V4 Pro+3.0Inside the noise
vs Claude Opus 4.8+4.8Clearly 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