Can Large Language Models Handle Discourse Par: leaderboard
Discourse particles, such as well and kind of, are crucial components that enable LLMs to "speak" more like humans. They are used to convey emotions, intentions, and interpersonal meanings. However, existing studies have not yet built a comprehensive understanding of LLMs' capabilities in handling discourse particles. Moreover, the limited number of studies focuses primarily
Metric: Overall Avg (MS) (self-reported). Source: benchmarklist.com. Saturation forecast: Around August 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 72.4 | #98 |
| 2 | DeepSeek V4 Flash | 69.8 | #137 |
| 3 | DeepSeek V4 Pro | 69.6 | #102 |
| 4 | Gemini 3.1 Pro (Preview) | 69.1 | #58 |
| 5 | Claude Sonnet 4.6 | 64.4 | #90 |
| 6 | Claude Haiku 4.5 | 64.4 | #280 |
| 7 | GPT-5.4 Mini | 62.8 | #212 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=can-large-language-models-handle-discourse-par · How It Works · Data refreshed daily, snapshot 2026-10-20.