CoreCognition Image + Video Boundary: leaderboard

CoreCognition Image + Video Boundary — 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
1GPT-4o87.17#744
2llava-video-7b-qwen2_multi_frameCoreCognition run: native-setting=llava-video-7b-qwen2_multi_frame86.52
3Ovis1.6-Gemma2-9B85.39#2314
4LLaVA-LLaMA-3-8B85.11#7395
5Qwen 2 VL 7B Instruct83.99#2596
6Claude 3.7 Sonnet (20250219)82.87#414
7DeepSeek-VL2-Small82.68#2401
8Gemini 1.5 Pro82.58#1061
9llava-video-7b-qwen2-video-only_multi_frameCoreCognition run: native-setting=llava-video-7b-qwen2-video-only_multi_frame82.49
10eagle-x5-13B-chat82.49
11Qwen-VL-Max-2025-01-2582.4
12Qwen 2.5 VL 7B Instruct82.3#1867
13internvl-chat-v1-2-plus82.3
14internvl-chat-v1-282.02
15DeepSeek-VL-7B-Chat81.84#3413

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