CoreCognition Single Image Hierarchy: leaderboard

CoreCognition Single Image Hierarchy — 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
1qwen2.5-vl-72b-instruct_videoCoreCognition run: native-setting=qwen2.5-vl-72b-instruct_video68.45
2LLaVA-Video-72B-Qwen266.1#1228
3yi-vision-v264.64
4qwen2.5-vl-7b-instruct_videoCoreCognition run: native-setting=qwen2.5-vl-7b-instruct_video62.38
5VideoLLaMA2-72B61.73#1355
6qwen2.5-vl-3b-instruct_videoCoreCognition run: native-setting=qwen2.5-vl-3b-instruct_video60.84
7InternVL2.5-78B60.52#1044
8Llama 3.2 90B Vision Instruct59.87#1769
9InternVL2.5-26B59.87#2269
10LLaVA-Video-7B-Qwen259.87#1854
11LLaVA-CoT57.69#3271
12InternVL2.5-8B56.15#2444
13VideoLLaMA2.1-7B-16F55.99#1924
14InternVL2.5-4B53.8#2583
15videochat2-stage3-mistral-7B51.62

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-hierarchy · How It Works · Data refreshed daily, snapshot 2026-11-16.