KWBench - Mean Score: leaderboard

Mean gated rubric score on KWBench (Knowledge Work Bench, 2026): 223 practitioner-sourced knowledge-work tasks in which the model must recognize, unprompted, the game-theoretic structure of a professional situation from raw inputs; a Gemini 3 Flash judge checks rubric criteria, best of three runs per model; a task scores 0 when a mandatory criterion fails and otherwise earns credit for further criteria. Percent; higher is better.

Metric: Mean task score (%): gated rubric score (0 when a mandatory criterion fails, otherwise 0.4 plus credit for good-to-have and ideal criteria) averaged over all evaluated tasks, over KWBench's 223 practitioner-sourced knowledge-work tasks (acquisitions, negotiations, clinical pharmacy, organizational politics, fraud analysis, incentive design), each encoding a game-theoretic pattern the model must recognize unprompted from raw inputs, with a code interpreter; a Gemini 3 Flash judge checks 15 binary rubric criteria per task, any failed mandatory criterion scores the task zero; best of three runs per model at provider-recommended temperature; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.622.6#113
2GPT-5.417.7#135
3GLM-5-Turbo16.8#255
4Qwen 3.5 Plus12.1#242
5Gemini 3.1 Pro (Preview)12#99
6Kimi K2.511.3#266
7KWBench Qwen 3.5 MoE (checkpoint unspecified)11.3
8MiniMax-M2.711.1#539
9Gemini 3 Flash10.7#175
10MiniMax-M2.58.5#600
11GPT-OSS-120B8.1#704

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=kwbench-mean-score · How It Works · Data refreshed daily, snapshot 2026-11-21.