ParaIntent - Response Quality: leaderboard
ParaIntent (2026) is a Chinese spoken emotional dialogue benchmark of 14 intent categories with balanced explicit and implicit samples, where a model hears a user utterance and returns a spoken response. Response quality is scored on the synthetic test set where a held-out DeepSeek-V4-Pro judge scores content, tone appropriateness and naturalness, in percent, higher is better.
Metric: Response quality score (%; held-out DeepSeek-V4-Pro judge of content, tone appropriateness and naturalness; over the ParaIntent synthetic test set (14 Chinese spoken-emotional-dialogue intent categories, balanced explicit and implicit samples); single-turn spoken response). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3-Omni | 58.65 | #999 |
| 2 | MiMo-Audio | 52.91 | #893 |
| 3 | Step-Audio-2-mini | 45.87 | |
| 4 | GLM-4-Voice | 45.18 | |
| 5 | Qwen2.5-Omni | 35.68 | #1499 |
Interactive version: theaggregate.ai/benchmark?slug=paraintent-response-quality · How It Works · Data refreshed daily, snapshot 2026-11-12.