DailyReport (Claude Code) - SubTask Pass: leaderboard
Metric: Share of subtasks (0-1 scaled to %) that satisfy every rubric criterion; open-ended daily search tasks on trending topics decomposed into subtasks with cascade rubrics, judged by Gemini 3 Flash; LLMs as backends of the Claude Code agent with Serper search and Jina Reader over MCP, extended thinking enabled; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (Medium) | 47.8 |
| 2 | GLM-5 | 26.5 |
| 3 | Gemini 3.1 Pro (Preview) | 26.2 |
| 4 | Kimi K2.5 | 22.3 |
| 5 | Claude Opus 4.6 (Thinking) | 20.6 |
Interactive version: theaggregate.ai/benchmark?slug=dailyreport-claude-code-subtask-pass · How It Works · Data refreshed daily, snapshot 2026-09-29.