ArgBench (Zero-Shot) - Perspective Assessment: leaderboard

ArgBench (2026) unifies 46 computational-argumentation tasks from 33 existing datasets into five skills. This board is the macro F1 averaged over the perspective assessment tasks under zero-shot prompting, times 100; higher is better.

Metric: Macro F1 (0-1, times 100) averaged over the argument perspective assessment (stance, framing and similar) tasks, zero-shot prompting; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-4.160.7#505
2Mixtral 8x7B Instruct55.6#4696
3Llama 3.3 70B Instruct55.2#1385
4Qwen 3 32B53.3#972
5Llama 3.1 8B Instruct48.6#4600
6Qwen 3 4B47.6#2348
7Phi-mini-MoE-instruct41
8Mistral-Small-Instruct-240940.9#2016
9DeepSeek R1 Distill Qwen 32B36.2#1834
10Mistral 7B Instruct (v0.3)33.1#6750
11DeepSeek-R1-Distill-Qwen-7B32.9#8055

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=argbench-zero-shot-perspective-assessment · How It Works · Data refreshed daily, snapshot 2026-11-21.