SOCO - Visual and Text Query: leaderboard

SOCO (2026) tests whether models can match corresponding object parts across images, here posed to vision-language models as four-choice questions over candidate keypoints in a target image, scored strictly so that a question counts only if answered correctly under all four answer orders. In this board the query keypoint is marked on a source image and also described in words; accuracy in percent, higher is better.

Metric: Accuracy (%) when the query keypoint is marked on the source image and also described in text, four-choice semantic-correspondence questions on SOCO image pairs (target image with candidate keypoints A to D), CircularEval: correct only if the model picks the right keypoint under all four answer permutations (random 0.4, consistent-permutation baseline 25); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1InternVL3.5-8B38.5#1618
2GPT-4o30.9#754
3Qwen 2.5 VL 7B Instruct30.8#1919
4Qwen 3 VL 8B30.8#927
5Qwen 3 VL 4B18#1163
6Qwen 2.5 VL 3B Instruct17.4#2720
7LLaVA-OneVision-7B14.1#3002

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=soco-visual-and-text-query · How It Works · Data refreshed daily, snapshot 2026-11-22.