CoreCognition Single Image Intentionality: leaderboard

CoreCognition Single Image Intentionality — 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: Estimated already saturated. 33 models tracked. Updated 16 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1VideoLLaMA2-72B85.83#1355
2Llama 3.2 90B Vision Instruct84.58#1769
3qwen2.5-vl-72b-instruct_videoCoreCognition run: native-setting=qwen2.5-vl-72b-instruct_video80.45
4LLaVA-Video-72B-Qwen280.06#1228
5LLaVA-Video-7B-Qwen278.53#1854
6qwen2.5-vl-3b-instruct_videoCoreCognition run: native-setting=qwen2.5-vl-3b-instruct_video77.65
7llava-next-video-32B-qwen77.63
8InternVL2.5-38B76.87#1413
9InternVL2.5-26B76.87#2269
10qwen2.5-vl-7b-instruct_videoCoreCognition run: native-setting=qwen2.5-vl-7b-instruct_video76.4
11yi-vision-v275.93
12InternVL2.5-78B75#1044
13InternVL2.5-8B72.66#2444
14VideoLLaMA2.1-7B-16F72.2#1924
15videollama2-7B-16f68.46

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

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