GPT-6.1 Sol

OpenAI, closed weights, released Sep 2026, run at provider defaults through the OpenAI 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

65.4

95% interval 55.7–73.9

Rank: 3of 14 models by Index; the paired intervals allow 1 to 9
Cost per page: $0.15Generation only, list price on run day
Second judge: 79.8GPT-5.6 Terra places it third of 14. The Claude judge: third.
Grade sum: 25.30Seven 1–5 grades, out of 35
Pages judged: 1010 judge passes, 2 builds per brief
Defects per page: 5.2Text under 12px (10 of 10)

Rank 3 of 14; the data allows anywhere from 1 to 9. Within noise of GPT-6 Astra, Grok 4.7, Claude Sonnet 5.5, GPT-6 Luna, Gemini 4 Argon, Muse Wonder, Claude Opus 5.5 and GPT-5.6 Sol: the gap to each sits inside the noise, so the board can't order them yet. Not clearly behind any model. Clearly ahead of Claude Fable 5.1, Claude Fable 5, Gemini 3.8 Flash, DeepSeek V4 Pro and Claude Opus 4.8.

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

Hierarchy3.802 of 14
Typography3.652 of 14
Spacing3.253 of 14
Color3.702 of 14
Components3.603 of 14
Composure3.605 of 14
Craft3.702 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.

Landing70.52 of 14
Dashboard68.81 of 14
Settings47.313 of 14
Pricing61.63 of 14
Article78.62 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: 102sMedian over every build, broken ones included
Output tokens: 7.2kMedian per page, reasoning included
Reasoning tokens: 0.4kMedian per page, as the API reports them
Broken pages: 0 of 10Every build came back as a whole page
Criticals: 80%8 of 10 scored pages had a critical issue in their first Rams code review
Built through: OpenAI APIServed by OpenAI

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: Article78.6
Index on this page, one judge pass. Brief mean 78.6.
- Aside note sits visually orphaned below TOCModerate
- Byline elements optically misaligned verticallyModerate
- Eyebrow label and issue number gap feels arbitraryModerate
Worst: Settings46.4
Index on this page, one judge pass. Brief mean 47.3.
- Mismatched button heights in avatar actions rowSerious
- Mismatched button weights in avatar action rowSerious
- 'Save changes' CTA placed before content it controlsSerious

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

Head to head

GPT-6.1 Sol against every other model: the paired difference in Index points, and whether the 95% interval on it clears zero.

vs GPT-6 Astra-1.1Inside the noise
vs Grok 4.7-0.7Inside the noise
vs Claude Sonnet 5.5+3.6Inside the noise
vs GPT-6 Luna+3.6Inside the noise
vs Gemini 4 Argon+3.8Inside the noise
vs Muse Wonder+4.8Inside the noise
vs Claude Opus 5.5+5.0Inside the noise
vs GPT-5.6 Sol+6.4Inside the noise
vs Claude Fable 5.1+6.4Clearly better
vs Claude Fable 5+7.8Clearly better
vs Gemini 3.8 Flash+9.8Clearly better
vs DeepSeek V4 Pro+10.7Clearly better
vs Claude Opus 4.8+12.5Clearly 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