Hedge-Bench - Valid Move Coverage: leaderboard
Metric: Valid moves covered (%), macro-averaged over tasks: mean within-theme share of expert moves hit, after discounting moves supported by hallucinated facts; 102 open-ended financial-analyst research tasks derived from recorded expert analyst discussions, run in the Harbor harness with the Terminus 2 agent over a sandboxed document corpus, eight trials per task, graded by a Gemini-3.1-Pro judge against expert rubric moves with hallucination-supported moves discounted; higher is better. Source: arxiv.org. Saturation forecast: Around November 2027. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 54.9 |
| 2 | Claude Opus 4.7 | 53.9 |
| 3 | Gemini 3.5 Flash | 49.8 |
| 4 | GPT-5.5 | 49.7 |
| 5 | Claude Opus 4.8 | 49.1 |
| 6 | Gemini 3.1 Pro (Preview) | 37.6 |
| 7 | Claude Haiku 4.5 | 36.9 |
| 8 | GPT-5.4 Mini | 26.1 |
Interactive version: theaggregate.ai/benchmark?slug=hedge-bench-valid-move-coverage · How It Works · Data refreshed daily, snapshot 2026-09-29.