CoreCognition Image only Permanence: leaderboard

CoreCognition Image only Permanence — 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
1points-qwen-2.5-7B-chat100
2Step-1.5V-mini93.75#2720
3step-1v-32k87.5
4omchat87.5
5GLM-4V-9B84.38#3333
6GLM-4V78.13#3924
7cogvlm2-llama3-chat-19B78.13
8GLM-4V-Plus75#1400
9Cambrian-13B75#7053
10wemm-chat-2k-cn75
11Molmo 7B-O71.88#6979
12Molmo-7B-D-092471.88#7502
13Molmo-72B-092471.88#5272
14step-1o-turbo-vision71.88
15LLaVA-v1.5-13B68.75#7610

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