CoreCognition Image only Mean: leaderboard

CoreCognition Image only Mean — CoreCognition native component accuracy and reported mean. Image+Video, image-only and single-image populations are separate; no pooling or replacement across cohorts. Metric: Accuracy (%). Source: https://grow-ai-like-a-child.github.io/CoreCognition/leaderboard

Metric: Accuracy (%). Source: grow-ai-like-a-child.github.io. Saturation forecast: Rough model projection: around 2026. 77 models tracked. Updated 16 Nov 2026.

Top models

#ModelScoreOverall rank
1Step-1.5V-mini69.62#2720
2step-1o-turbo-vision69.49
3GLM-4V-Plus68.22#1400
4points-qwen-2.5-7B-chat66.5
5NVLM-D-72B66.03#2181
6wemm-chat-2k-cn65.79
7omchat65.43
8step-1v-32k64.5
9GLM-4V-9B64.17#3333
10Molmo-72B-092463.96#5272
11Cambrian-34B63.08#5295
12Cambrian-13B62.79#7053
13InternLM-XComposer2-4KHD61.36#6477
14Ovis1.6-Gemma2-27B61.3#1833
15WeMM61.01#5473

Interactive version: theaggregate.ai/benchmark?slug=corecognition-image-only-mean · How It Works · Data refreshed daily, snapshot 2026-11-16.