RefGlitch-Bench: leaderboard

RefGlitch-Bench (2026) tests whether vision-language models spot visual glitches in video game frames when shown a clean reference frame of the same scene. This board is accuracy from 0 to 1 for binary glitch detection on a balanced set of glitchy and glitch-free frames, each shown with a manually selected clean reference frame; higher is better.

Metric: Accuracy (fraction) of binary gameplay-glitch detection on 1,000 balanced test frames (500 glitchy, 500 glitch-free) with a manually selected clean reference frame (Ref setting); higher is better. Source: arxiv.org. Saturation forecast: Around January 2028. 5 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-50.77#178
2Qwen 3 VL 8B0.73#917
3Ministral 3 8B0.68#2131
4Gemini 3 Flash (Preview)0.63#152
5Gemma 3 12B0.63#2059

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=refglitch-bench · How It Works · Data refreshed daily, snapshot 2026-11-21.