HalluScore - Adversarial: leaderboard
HalluScore (2026) is an Arabic question-answering benchmark of 827 hallucination-prone Modern Standard Arabic questions, with human annotation of model answers. This board is the factual hallucination rate, the percent of answers that state an unsupported fact with confidence on the 360 adversarial questions with misleading phrasing, zero-shot at temperature 0 (expressed uncertainty or refusal does not count); lower is better.
Metric: Factual hallucination rate (%) on the 360 adversarial HalluScore questions (misleading or carefully crafted phrasing), 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 February 2027. 17 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5 | 31.11 | #178 | |
| 2 | Claude Opus 4 | 34.72 | #326 | |
| 3 | Claude Sonnet 4.5 | 38.89 | #261 | |
| 4 | GPT-4o | 44.72 | #751 | |
| 5 | O4 Mini | 51.11 | #368 | |
| 6 | Qwen3-235B-A22B-Instruct-2507-tput | 53.61 | #548 | |
| 7 | Grok 4 Fast | Reasoning | 55.56 | #447 |
| 8 | Qwen 3 Next 80B A3B Instruct | 58.61 | #778 | |
| 9 | DeepSeek R1 | 60.83 | #482 | |
| 10 | DeepSeek V3 | 62.5 | #729 | |
| 11 | Grok 4 | 65.28 | #319 | |
| 12 | Llama 4 Maverick Instruct FP8 | 70.56 | #1023 | |
| 13 | ALLaM-7B-Instruct-preview | 75 | #2001 | |
| 14 | Fanar-1-9B | 80.83 | #1913 | |
| 15 | O3 | 81.67 | #237 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=halluscore-adversarial · How It Works · Data refreshed daily, snapshot 2026-11-21.