CoreCognition Single Image Continuity: leaderboard
CoreCognition Single Image Continuity — 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
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | videochat2_hd_stage4_mistral_7b_hf | 70.64 | ||
| 2 | videochat2-stage3-mistral-7B | 69.19 | ||
| 3 | qwen2.5-vl-72b-instruct_video | CoreCognition run: native-setting=qwen2.5-vl-72b-instruct_video | 66.86 | |
| 4 | InternVL2.5-38B | 66.57 | #1413 | |
| 5 | InternVL2.5-26B | 66.57 | #2269 | |
| 6 | VideoLLaMA2.1-7B-16F | 66.28 | #1924 | |
| 7 | llava-next-video-32B-qwen | 66.09 | ||
| 8 | LLaVA-Video-7B-Qwen2 | 63.79 | #1854 | |
| 9 | yi-vision-v2 | 63.08 | ||
| 10 | videollama2-7B-16f | 62.79 | ||
| 11 | Llama 3.2 11B Instruct | 61.63 | #5235 | |
| 12 | Llama 3.2 90B Vision Instruct | 61.63 | #1769 | |
| 13 | VideoLLaMA2-72B | 61.63 | #1355 | |
| 14 | qwen2.5-vl-3b-instruct_video | CoreCognition run: native-setting=qwen2.5-vl-3b-instruct_video | 61.05 | |
| 15 | qwen2.5-vl-7b-instruct_video | CoreCognition run: native-setting=qwen2.5-vl-7b-instruct_video | 60.76 |
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-continuity · How It Works · Data refreshed daily, snapshot 2026-11-16.