CoreCognition Image + Video Intuitivephysics: leaderboard

CoreCognition Image + Video Intuitivephysics — 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. 120 models tracked. Updated 16 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1O1 (2024-12-17)75.45#313
2Idefics3-8B-Llama365.3#5972
3GPT-4 Turbo63.48#999
4Qwen 2.5 VL 3B Instruct62.12#2562
5Qwen 2 VL 7B Instruct61.82#2596
6Ovis1.6-Llama3.2-3B61.36#3459
7internvl-chat-v1-161.06
8VILA1.5-13B60.76#7184
9Phi-3.5-Vision60.61#4877
10llama-3-vila1.5-8b-fixCoreCognition run60.61
11Qwen 2 VL 2B Instruct60.45#4986
12InternVL2-76B60.15#1519
13Mini-InternVL-Chat-4B-V1.560#4881
14IDEFICS-80B-Instruct59.85#8917
15DeepSeek-VL2-Small59.85#2401

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

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