CAREBench - Positive Emotion Labels: leaderboard
Metric: Example-F1 (%; per-narrative F1 of the selected positive emotion labels against the narrator's own labels; 1,000 narratives, zero-shot, mean of five runs for open models). Source: arxiv.org. Saturation forecast: Around 2031. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 (Non-reasoning) | 55.6 |
| 2 | Qwen 3.5 9B (Non-reasoning) | 51.3 |
| 3 | Claude Sonnet 4.6 (Non-reasoning) | 49 |
| 4 | Llama 3.1 8B Instruct | 45 |
Interactive version: theaggregate.ai/benchmark?slug=carebench-positive-emotion-labels · How It Works · Data refreshed daily, snapshot 2026-09-26.