ArgBench (Zero-Shot) - Quality Assessment: leaderboard
Metric: Macro F1 (0-1, times 100) averaged over the argument quality assessment tasks, zero-shot prompting; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 | 44 |
| 2 | Llama 3.3 70B Instruct | 41.9 |
| 3 | Qwen 3 32B | 39.4 |
| 4 | DeepSeek-R1-Distill-Qwen-7B | 38.3 |
| 5 | Llama 3.1 8B Instruct | 37.7 |
| 6 | DeepSeek R1 Distill Qwen 32B | 37.3 |
| 7 | Mistral 7B Instruct (v0.3) | 36 |
| 8 | Mistral-Small-Instruct-2409 | 33.5 |
| 9 | Qwen 3 4B | 33.2 |
| 10 | Mixtral 8x7B Instruct (v0.1) | 33.2 |
Interactive version: theaggregate.ai/benchmark?slug=argbench-zero-shot-quality-assessment · How It Works · Data refreshed daily, snapshot 2026-10-07.