Which AI model scores highest on AA-Briefcase v1.1 (rubric)?

Claude Fable 5.2 by Anthropic holds the best AA-Briefcase v1.1 (rubric) result among tracked models, at 66.1% (released Nov 9 2026). Higher scores are better on this benchmark.

What are the top 5 models on AA-Briefcase v1.1 (rubric)?

1. Claude Fable 5.2 (Anthropic) — 66.1%; 2. Gemini 4 Carbon (Google) — 61.8%; 3. Claude Fable 5.1 (Anthropic) — 61.5%; 4. Claude Opus 5.5 (Anthropic) — 61.2%; 5. GPT-6 Astra (OpenAI) — 52%.

How many models have a published AA-Briefcase v1.1 (rubric) score?

5 tracked models have a published AA-Briefcase v1.1 (rubric) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

Knowledge work

AA-Briefcase v1.1 (rubric)

Artificial Analysis agentic office-work eval, v1.1 rubric score; separate from the Elo rating

Rankings

Higher is better

AA-Briefcase v1.1 (rubric) — frequently asked questions

: Artificial Analysis agentic office-work eval, v1.1 rubric score; separate from the Elo rating
: Claude Fable 5.2 by Anthropic holds the best AA-Briefcase v1.1 (rubric) result among tracked models, at 66.1% (released Nov 9 2026). Higher scores are better on this benchmark.
: 1. Claude Fable 5.2 (Anthropic) — 66.1%; 2. Gemini 4 Carbon (Google) — 61.8%; 3. Claude Fable 5.1 (Anthropic) — 61.5%; 4. Claude Opus 5.5 (Anthropic) — 61.2%; 5. GPT-6 Astra (OpenAI) — 52%.
: 5 tracked models have a published AA-Briefcase v1.1 (rubric) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.