Kev Release - tasksource-heldout-v1 Test Accuracy (17 Families): leaderboard
2024 Kev-4B tasksource values are development only. Metric: accuracy (+ ECE secondary).
Metric: accuracy (+ ECE secondary). Source: huggingface.co. Saturation forecast: Rough model projection: around 2026. 2 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kev 27B | 0.795 | |
| 2 | Kev 27B v1 | 0.743 |
Interactive version: theaggregate.ai/benchmark?slug=kev-release-tasksource-heldout-v1-test-accuracy-17-families · How It Works · Data refreshed daily, snapshot 2026-11-16.