SPACE Personality Study - Base - FOMC - Macro-F1: leaderboard
SPACE v1 Table11,FOMC unmodified Instruct Base only, nativeMacro-F1. Base/T/F triple retained inprimary; takeBasefirstonly. Greedy and exactstudyextraction. CommitmentBankstrictmajority858/1009records,151distributiononly,notoriginal3classgold. No DPOrow substitution.
Metric: Macro-F1 (native). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 7B Instruct | 0.57 | #2583 |
| 2 | Llama 3.1 8B Instruct | 0.48 | #4277 |
| 3 | Qwen 2.5 3B Instruct | 0.44 | #5206 |
| 4 | Llama 3.2 3B Instruct | 0.4 | #7185 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=space-personality-study-base-fomc-macro-f1 · How It Works · Data refreshed daily, snapshot 2026-11-12.