CoreCognition Image only Toolusing: leaderboard

CoreCognition Image only Toolusing — 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
1Ovis1.6-Gemma2-27B100#1833
2NVLM-D-72B100#2181
3GLM-4V-Plus100#1400
4Yi-VL-34B100#6990
5Bunny-Llama3-8B100#7560
6WeMM100#5473
7InternLM-XComposer2-4KHD100#6477
8Cambrian-34B100#5295
9Reka Core100#3349
10Molmo-72B-0924100#5272
11wemm-chat-2k-cn100
12llava-internlm-7B100
13step-1o-turbo-vision99.17
14Cambrian-8B98.96#6834
15step-1v-32k98.96

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