ReactBench (Hallucination) - Counterfactual Attribute (Standard): leaderboard
Accuracy in percent on ReactBench (2026) questions about images whose object attribute was edited against expectation, with standard prompting; an incorrect answer counts as a hallucination. Higher is better.
Metric: Accuracy (%) on Counterfactual Attribute (7,668 questions on 639 images with an edited attribute) with standard prompting, hallucination-inducing questions (multiple choice, short answer, yes/no) about images edited by Qwen-Image-Edit-2511; an incorrect answer counts as a hallucination; temperature 0.2, at most 1,024 tokens, LLM answer matching; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 27B | 56.3 | #375 |
| 2 | Qwen 3 VL 32B Instruct | 55.4 | #606 |
| 3 | InternVL2.5-26B | 54.6 | #2156 |
| 4 | InternVL3-8B | 54.4 | #1782 |
| 5 | Qwen 3 VL 4B | 53.5 | #1151 |
| 6 | Qwen 2.5 VL 72B Instruct | 53.4 | #841 |
| 7 | InternVL2.5-8B | 50.7 | #2370 |
| 8 | Qwen 2.5 VL 7B Instruct | 49.2 | #1902 |
| 9 | Qwen 3.5 9B | 48.7 | #828 |
| 10 | ReactBench LLaVA-OV-7B (checkpoint unspecified) | 47.7 | |
| 11 | ReactBench LLaVA-v1.6-7B (checkpoint unspecified) | 35.4 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=reactbench-hallucination-counterfactual-attribute-standard · How It Works · Data refreshed daily, snapshot 2026-11-21.