ReactBench (Hallucination) - Alteration Tracing (Standard): leaderboard
Accuracy in percent on ReactBench (2026) questions about before-and-after image pairs, asking what changed, with standard prompting; an incorrect answer counts as a hallucination. Higher is better.
Metric: Accuracy (%) on Alteration Tracing (10,956 questions on 913 before-and-after image pairs) 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 February 2027. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 27B | 79.9 | #380 |
| 2 | Qwen 3 VL 32B Instruct | 73.5 | #607 |
| 3 | Qwen 3.5 9B | 67.4 | #838 |
| 4 | InternVL3-8B | 64.1 | #1795 |
| 5 | Qwen 2.5 VL 72B Instruct | 62.1 | #849 |
| 6 | Qwen 3 VL 4B | 58 | #1163 |
| 7 | Qwen 2.5 VL 7B Instruct | 54 | #1919 |
| 8 | InternVL2.5-26B | 51.3 | #2174 |
| 9 | InternVL2.5-8B | 45.4 | #2393 |
| 10 | ReactBench LLaVA-OV-7B (checkpoint unspecified) | 30.6 | |
| 11 | ReactBench LLaVA-v1.6-7B (checkpoint unspecified) | 27.3 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=reactbench-hallucination-alteration-tracing-standard · How It Works · Data refreshed daily, snapshot 2026-11-22.