HELM Classic - CNN/DailyMail: leaderboard
The "CNN/DailyMail" subtask of HELM Classic. A scenario from Stanford HELM Classic, the holistic evaluation suite scoring language models across accuracy, robustness and calibration on a broad set of tasks.
Metric: ROUGE-2 (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 42 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | text-davinci-003 | 15.59 | #1146 |
| 2 | GLM-130B | 15.44 | #1568 |
| 3 | text-davinci-002 | 15.34 | #1225 |
| 4 | text-curie-001 | 15.16 | #1632 |
| 5 | text-babbage-001 | 15.11 | #1648 |
| 6 | OPT-175B | 14.59 | #1538 |
| 7 | OPT-66B | 13.6 | #1603 |
| 8 | text-ada-001 | 13.56 | #1655 |
| 9 | gpt-j-6B | 13.11 | #1609 |
| 10 | davinci | 12.71 | #1553 |
| 11 | gpt-neox-20B | 12.3 | #1600 |
| 12 | Palmyra X (43B) | 4.91 | #1317 |
| 13 | T5 (11B) | 4.28 | #1652 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-cnn-dailymail · How It Works · Data refreshed daily, snapshot 2026-10-20.