OpenEval - CNN/DailyMail — leaderboard
Metric: ROUGE-L. Source: huggingface.co. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B | 25.62 |
| 2 | Llama 2 7B | 25.28 |
| 3 | Llama 2 13B | 25.26 |
| 4 | LLaMA-30B | 24.86 |
| 5 | LLaMA-13B | 24.2 |
| 6 | LLaMA-7B | 23.52 |
| 7 | LLaMA-65B | 23.18 |
| 8 | falcon-7B | 23.09 |
| 9 | RedPajama-INCITE-Instruct-3B-v1 | 21.74 |
| 10 | pythia-6.9B | 21.59 |
| 11 | falcon-7B Instruct | 21.43 |
| 12 | RedPajama-INCITE-Base-3B-v1 | 19.88 |
| 13 | vicuna-13B-v1.3 | 17.59 |
| 14 | vicuna-7B-v1.3 | 0.46 |
| 15 | alpaca-7B | 0.21 |
Interactive version: theaggregate.ai/benchmark?slug=openeval-cnn-dailymail · How the rankings work · Data refreshed daily, snapshot 2026-07-22.