SeaEval - Dialogue - SAMSum (Zero-Shot) — leaderboard

Metric: Average ROUGE (0-100). Source: huggingface.co. 25 models tracked.

Top models

#ModelScore
1Gemma 2 2B (IT)31.12
2Gemma 2 9B (IT)31.01
3Llama 3 8B Cpt Sea Lionv2.1 Instruct30.5
4Qwen 2.5 7B Instruct29.88
5SeaLLMs-v3-7B-Chat29.6
6Llama 3 70B Instruct28.94
7Llama 3.1 70B Instruct28.93
8Qwen 2.5 72B Instruct28.85
9Llama 3 8B Instruct28.46
10Qwen 2.5 32B Instruct28.44
11Llama 3.1 8B Instruct28.21
12Qwen 2 72B Instruct28.01
13Qwen 2.5 14B Instruct27.14
14Qwen 2.5 3B Instruct26.94
15Qwen 2 7B Instruct25.67

Interactive version: theaggregate.ai/benchmark?slug=seaeval-dialogue-samsum-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.