CoreCognition Image only Toolusing: leaderboard
CoreCognition Image only 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: Rough model projection: around 2026. 77 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ovis1.6-Gemma2-27B | 100 | #1833 |
| 2 | NVLM-D-72B | 100 | #2181 |
| 3 | GLM-4V-Plus | 100 | #1400 |
| 4 | Yi-VL-34B | 100 | #6990 |
| 5 | Bunny-Llama3-8B | 100 | #7560 |
| 6 | WeMM | 100 | #5473 |
| 7 | InternLM-XComposer2-4KHD | 100 | #6477 |
| 8 | Cambrian-34B | 100 | #5295 |
| 9 | Reka Core | 100 | #3349 |
| 10 | Molmo-72B-0924 | 100 | #5272 |
| 11 | wemm-chat-2k-cn | 100 | |
| 12 | llava-internlm-7B | 100 | |
| 13 | step-1o-turbo-vision | 99.17 | |
| 14 | Cambrian-8B | 98.96 | #6834 |
| 15 | step-1v-32k | 98.96 |
Interactive version: theaggregate.ai/benchmark?slug=corecognition-image-only-toolusing · How It Works · Data refreshed daily, snapshot 2026-11-16.