AA-Briefcase - PPTX - Rubric Pass Rate: leaderboard

Artificial Analysis's private AA-Briefcase business-workflow evaluation restricted to tasks whose deliverable is a PowerPoint (PPTX) file: the share of rubric criteria the deliverables pass. Reported 0-100 (%); higher is better. Task-weighted, these file-type rates reproduce the AA-Briefcase rubric pass rate.

Metric: Rubric Pass Rate (%). Source: artificialanalysis.ai. Saturation forecast: Around July 2027. 249 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 4 Argon (High)67.36#9 (Gemini 4 Argon)
2Gemini 4 Carbon (High)62.04#1 (Gemini 4 Carbon)
3Muse Wonder Full (Max)57.14#12 (Muse Wonder Full)
4Muse Spark 1.3 (Max)56.73#31 (Muse Spark 1.3)
5Muse Wonder (Max)55.51#18 (Muse Wonder)
6Qwen3.8-Flash-Next55.1#76
7Qwen 3.8 Max (0902)53.88#36
8Claude Opus 5 (Max)53.47#16 (Claude Opus 5)
9Grok 4.6 (xHigh)52.65#29 (Grok 4.6)
10Claude Haiku 5.5 (Max)52.24#43 (Claude Haiku 5.5)
11MiMo-V2.6-Pro51.84#48
12Claude Opus 5 (High)51.84#16 (Claude Opus 5)
13Grok 4.7 (xHigh)51.43#26 (Grok 4.7)
14Grok 4.6 (High)51.02#29 (Grok 4.6)
15Grok 4.7 (High)50.61#26 (Grok 4.7)

No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.

Interactive version: theaggregate.ai/benchmark?slug=aa-briefcase-pptx-rubric-pass-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.