CommunityFact (Closed-Input) - Finance (Spanish): leaderboard
Metric: Macro-F1 (%) over True and False claims on the temporally held-out Finance Spanish test slice (82 True and 52 False standalone claims built from helpful X Community Notes, labels grounded in the note), closed-input (the model sees only the claim and its timestamp, no web search), zero-shot verdict prompt in the claim language; open models greedy, closed models at default settings with reasoning effort medium in their thinking runs; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 4.3 (Medium) | 72.37 |
| 2 | aya-expanse-32B | 69.89 |
| 3 | Qwen 3 32B (Non-reasoning) | 65.73 |
| 4 | aya-expanse-8B | 65.21 |
| 5 | Gemini 2.5 Flash (Non-reasoning) | 64.17 |
| 6 | Gemini 2.5 Flash (Medium) | 63.38 |
| 7 | Qwen 3 14B (Non-reasoning) | 61.94 |
| 8 | Qwen 3 32B | 60.18 |
| 9 | Qwen 3 14B | 59.26 |
| 10 | EuroLLM-22B-Instruct-2512 | 56.09 |
| 11 | GPT-5 Nano (Medium) | 51.29 |
| 12 | Grok 4.3 (Non-reasoning) | 50.33 |
| 13 | Ministral-8B-Instruct-2410 | 36.92 |
| 14 | GPT-5 Nano (Minimal) | 34.48 |
Interactive version: theaggregate.ai/benchmark?slug=communityfact-closed-input-finance-spanish · How It Works · Data refreshed daily, snapshot 2026-10-07.