LoomBench (default; Typed Decisions) - Hold: leaderboard
Mean source-normalized Hold objective score (0-100), 11 agents with this objective; Five agents, personas and objectives; identical 11 match designs, 26 rounds, 100-unit loom; higher is better. Interfaces are separate and random controls omitted.
Metric: Mean source-normalized Hold objective score (0-100), 11 agents with this objective; Five agents, personas and objectives; identical 11 match designs, 26 rounds, 100-unit loom; higher is better. Interfaces are separate and random controls omitted.. Source: raw.githubusercontent.com. Saturation forecast: Around August 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Luna | 52.27 | #52 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=loombench-default-typed-decisions-hold · How It Works · Data refreshed daily, snapshot 2026-10-16.