OfficePrism-420 - Hermes (Search) - NL-DB Pass: leaderboard

OfficePrism-420 (arXiv 2611.22262) single-domain enterprise agent evaluation (Hermes (Search) - NL-DB Pass) across 420 tasks in 7 business functions.

Metric: Joint NL-DB Pass Rate (%) over 420 single-domain enterprise tasks. Source: arxiv.org. Saturation forecast: Around February 2027. 14 models tracked. Updated 22 Nov 2026.

Top models

#ModelScoreOverall rank
1GLM-5.272.6#154
2Kimi K371.9#62
3Claude Opus 4.770#86
4Claude Opus 4.869.5#64
5Claude Sonnet 569#124
6DeepSeek V4 Pro68.1#201
7Qwen 3.6 Max66.9#192
8DeepSeek V4 Flash66.4#273
9GLM-5.164#200
10Qwen 3.7 Plus62.6#221
11Qwen 3.7 Max62.4#160
12Kimi K2.661.4#197
13GPT-5.560.5#56
14Qwen 3.6 Plus53.6#245

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=officeprism-420-hermes-search-nl-db-pass · How It Works · Data refreshed daily, snapshot 2026-11-22.