Clearing FoG - WikiWebQuestions IO Prompt Token F1: leaderboard
Clearing FoG compares question answering without access to a knowledge graph. Direct answers, chain-of-thought prompting and self-consistency are measured separately. Token F1 (%).
Metric: Token F1 (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3 | 42.36 | #324 |
| 2 | GPT-4o Mini | 40.31 | #593 |
| 3 | Qwen 2.5 7B Instruct | 28.24 | #862 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=clearing-fog-wikiwebquestions-io-prompt-token-f1 · How It Works · Data refreshed daily, snapshot 2026-10-29.