CoreCognition Single Image Toolusing: leaderboard
CoreCognition Single Image Toolusing — 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 | InternVL2.5-8B | 100 | #2444 | |
| 2 | LLaVA-Video-72B-Qwen2 | 100 | #1228 | |
| 3 | VideoLLaMA2-72B | 100 | #1355 | |
| 4 | qwen2.5-vl-72b-instruct_video | CoreCognition run: native-setting=qwen2.5-vl-72b-instruct_video | 100 | |
| 5 | Llama 3.2 90B Vision Instruct | 99.17 | #1769 | |
| 6 | InternVL2.5-38B | 99.17 | #1413 | |
| 7 | VideoLLaMA2.1-7B-16F | 98.96 | #1924 | |
| 8 | llava-next-video-32B-qwen | 98.93 | ||
| 9 | qwen2.5-vl-7b-instruct_video | CoreCognition run: native-setting=qwen2.5-vl-7b-instruct_video | 98.33 | |
| 10 | videollama2-7B-16f | 98.13 | ||
| 11 | LLaVA-Video-7B-Qwen2 | 97.86 | #1854 | |
| 12 | InternVL2.5-26B | 97.64 | #2269 | |
| 13 | qwen2.5-vl-3b-instruct_video | CoreCognition run: native-setting=qwen2.5-vl-3b-instruct_video | 97.5 | |
| 14 | yi-vision-v2 | 97.22 | ||
| 15 | llava-video-7B-qwen2-video-only | 96.17 |
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-toolusing · How It Works · Data refreshed daily, snapshot 2026-11-16.