HalluScore - Factual Hallucination: leaderboard
Metric: Factual hallucination rate (%): share of the 827 HalluScore questions (hallucination-prone Modern Standard Arabic questions across eleven types and many domains), zero-shot short-answer Arabic question answering at temperature 0 with no retrieval, responses annotated by humans; an answer that states an unsupported fact with confidence counts, an expressed uncertainty or refusal does not; lower is better. Source: arxiv.org. Saturation forecast: Around January 2027. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 25.15 |
| 2 | Claude Opus 4 | 33.01 |
| 3 | Claude Sonnet 4.5 | 35.79 |
| 4 | GPT-4o | 41.23 |
| 5 | O4 Mini | 46.43 |
| 6 | Grok 4 Fast (Reasoning) | 52.96 |
| 7 | DeepSeek V3 | 54.78 |
| 8 | Llama 4 Maverick Instruct FP8 | 57.56 |
| 9 | DeepSeek R1 | 57.68 |
| 10 | Qwen 3 Next 80B A3B Instruct | 61.43 |
| 11 | Grok 4 | 65.42 |
| 12 | ALLaM-7B-Instruct-preview | 68.68 |
| 13 | Fanar-1-9B | 79.32 |
| 14 | O3 | 80.05 |
Interactive version: theaggregate.ai/benchmark?slug=halluscore-factual-hallucination · How It Works · Data refreshed daily, snapshot 2026-10-07.