RebusBench (1-Shot): leaderboard

RebusBench (2026) tests vision-language models on 1,164 human-authored rebus puzzles, where the picture and its text layout are clues to an idiom or phrase. Exact match (%) after normalization, with one solved example in the prompt; higher is better.

Metric: Exact match: exact match accuracy (%) against the gold answer after lowercasing and stripping whitespace and special characters, on the 1,164 human-authored rebus puzzles of RebusBench, one in-context example in the text prompt; higher is better. Source: arxiv.org. Saturation forecast: Around February 2028. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 2.5 VL 32B Instruct7.39#1062
2Qwen 3 VL 8B6.36#917
3Qwen 2.5 VL 7B Instruct5.15#1902
4InternVL3.5-30B-A3B5.15#1267
5InternVL3.5-8B4.81#1604
6InternVL3.5-4B4.73#1930
7Qwen 3 VL 4B4.12#1151
8Qwen 2.5 VL 3B Instruct3.52#2713
9LLaVA-v1.5-7B1.2#7630

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

Interactive version: theaggregate.ai/benchmark?slug=rebusbench-1-shot · How It Works · Data refreshed daily, snapshot 2026-11-21.