WorkBuddy Bench (CodeBuddy Code) - Office: leaderboard

Metric: Score (0-100) on the 50 office data and file workflow tasks scored by a task-specific blend of rule and evidence-grounded judge scores, run in the CodeBuddy Code harness in think mode, mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 7 models tracked.

Top models

#ModelScore
1CodeBuddy Code + Claude Opus 4.8 (Thinking)82.37
2CodeBuddy Code + Hy382.08
3CodeBuddy Code + GPT-5.581.96
4CodeBuddy Code + GLM-5.279.6
5CodeBuddy Code + DeepSeek-V4-Pro (Thinking)79.11
6CodeBuddy Code + MiniMax-M378.28
7CodeBuddy Code + DeepSeek-V4-Flash (Thinking)77.47

Interactive version: theaggregate.ai/benchmark?slug=workbuddy-bench-codebuddy-code-office · How It Works · Data refreshed daily, snapshot 2026-09-29.