CoreCognition Image only Perspective: leaderboard

CoreCognition Image only Perspective — 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: Rough model projection: around 2026. 77 models tracked. Updated 16 Nov 2026.

Top models

#ModelScoreOverall rank
1TransCore-M51.71#8171
2llava-v1.6-vicuna-13B-hf49.83
3llava-v1.6-vicuna-7B-hf49.14
4InstructBLIP-Vicuna-13B47.6
5falcon-11B-vlm47.09
6MMAlaya46.46#9186
7Molmo 7B-O46.29#6979
8MiniGPT-4-v245.78
9PaliGemma-3B-mix-44845.55#6985
10cogvlm-chat45.03
11RekaEdge44.98#7648
12LLaVA-v1.5-7B44.69#7894
13Monkey44.63#8891
14MolmoE-1B44.35#8333
15Monkey-Chat43.95#7874

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