EYT-Bench (Nemotron-USA) - Anthropomorphic Interaction: leaderboard
Metric: Anthropomorphic interaction (colloquial, emotional, flow, flexibility, rhythm) (0-100): judge rubric of five sub-indicators scored 0, 0.5 or 1 per turn, summed to 0-5, turn weighted with the first two turns at 0.10 each and rescaled to 0-100; multi-turn dialogues of up to 10 turns with a persona-grounded Gemini 3.1 Pro user simulator and a Gemini 3.1 Pro judge (reasoning effort high, temperature 0); Nemotron-Personas-USA pool (100 dialogues; structured demographic personas). Source: arxiv.org. Saturation forecast: Estimated already saturated. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Seed 2.0 Pro | 96 |
| 2 | Claude Sonnet 4.6 | 95.9 |
| 3 | Claude Opus 4.7 | 94.9 |
| 4 | Seed 2.0 Lite | 94.2 |
| 5 | DeepSeek V4 Pro | 92.9 |
| 6 | Qwen 3.5 27B | 92.4 |
| 7 | Gemini 3.1 Pro (Preview) (Thinking) | 90.9 |
| 8 | Qwen 3.5 397B A17B | 90.7 |
| 9 | DeepSeek V4 Flash | 89.2 |
| 10 | Qwen 3.5 35B A3B | 88.2 |
| 11 | GPT-5.5 | 75.6 |
| 12 | Seed 2.0 Mini | 75.4 |
Interactive version: theaggregate.ai/benchmark?slug=eyt-bench-nemotron-usa-anthropomorphic-interaction · How It Works · Data refreshed daily, snapshot 2026-09-29.