Mistral Large 4

Mistral, closed weights, released Oct 2026, run at provider defaults through OpenRouter, served by Mistral. It built the same five pages as every model on season 0.6.1 of the Rams Index, and Rams graded each one at desktop and phone size.

Rams Index, 0 to 100

72.0

95% interval 65.2–74.8

Rank: 9of 26 models by Index; the paired intervals allow 6 to 12
Cost per page: $0.29Generation only, list price on run day
Earlier judges: 84.3Under the grade sum this board used until 10/28, GPT-5.6 Terra placed it seventh of 26 and the Claude judge sixth.
Grade sum: 28.27Seven 1–5 grades, out of 35
Pages judged: 1010 judge passes, 2 builds per brief
Defects per page: 6.6Text under 12px (10 of 10)

Rank 9 of 26; the data allows anywhere from 6 to 12. Within noise of GLM 5.3, Claude Opus 5.5, Claude Sonnet 5.5, GPT-6 Luna, Claude Haiku 5.5 and GPT-5.6 Sol: the gap to each sits inside the noise, so the board can't order them yet. Clearly behind GPT-6.1 Sol, GPT-6 Astra, Grok 4.7, Claude Fable 5.1 and GPT-6.1 Terra. Clearly ahead of Claude Fable 5, Kimi K3, Kimi K3.1 Pro, GLM 5.3 Prime, DeepSeek V4 Pro, Claude Opus 4.8, MiMo V2.6 Pro, Gemini 4 Argon, Gemini 3.8 Flash, MiniMax M3, Qwen 3.8 Max, Qwen 3.8 Max Prime, Muse Spark Open and Muse Wonder.

Checked against one designer · 63% agreementSeason 0.6.1New 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 26 models.

Hierarchy3.953 of 26
Typography3.854 of 26
Spacing3.1514 of 26
Color4.002 of 26
Components3.5015 of 26
Composure3.657 of 26
Craft3.755 of 26

Bars start at 2.8 and end at 4.6, 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 26.

Landing84.47 of 26
Dashboard43.116 of 26
Settings64.914 of 26
Pricing92.55 of 26
Article54.412 of 26

Bars run from 0 to 100.

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: 672sMedian over every build, broken ones included
Output tokens: 65.3kMedian per page, reasoning included
Reasoning tokens: 53.8kMedian per page, as the API reports them
Broken pages: 0 of 10Every build came back as a whole page
Criticals: 40%4 of 10 scored pages had a critical issue in their first Rams code review
Built through: OpenRouterServed by Mistral

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: Pricing95.4
Index on this page, one judge pass. Brief mean 92.5.
- Inconsistent price numeral scale between tiersModerate
- Free tier card top edge misaligned with Pro/Team due to missing recommended tagModerate
- Save 20% pill inside toggle button creates competing focal pointModerate
Worst: Dashboard29.2
Index on this page, one judge pass. Brief mean 43.1.
- Episode names render as 'undefined'Serious
- Stat numerals use monospaced/tabular figures inconsistently with label casingSerious
- X-axis date labels clipped at right edgeModerate