LiveFact - Classification Accuracy: leaderboard
Metric: Accuracy (%) in Classification Mode (Real, Fake or Ambiguous against the time-invariant label) with the evidence available three days after the event, LiveFact November 2025; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 235B A22B 2507 Instruct | 82.08 |
| 2 | GPT-OSS-120B | 81.81 |
| 3 | GPT-5.1 | 81.01 |
| 4 | GPT-5.2 | 77.32 |
| 5 | Qwen 3 30B A3B 2507 Instruct | 77 |
| 6 | GPT-4o (2024-08-06) | 73.98 |
| 7 | Qwen 3 32B | 71.86 |
| 8 | GPT-4o Mini (2024-07-18) | 71.22 |
| 9 | Qwen 3 8B | 69.83 |
| 10 | Llama 3.3 70B Instruct | 69.76 |
| 11 | GPT-OSS-20B | 67.42 |
| 12 | Qwen 3 4B 2507 Instruct | 64.8 |
| 13 | DeepSeek V3.1 | 63.73 |
| 14 | Kimi K2 (Thinking) | 54.21 |
| 15 | Llama 3.1 8B Instruct | 50.2 |
Interactive version: theaggregate.ai/benchmark?slug=livefact-classification-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-07.