DV-World - DV-Inter Interaction Success: leaderboard
Metric: Interaction success rate (%): 0.5 + 0.5 x (resolved turns minus inappropriate refusals) / (requests + 1), on ambiguous visualization requests with a GPT-5-mini user simulator, run in the DV-World-Agent framework (data manipulation, multimodal perception and proactive interaction tools); expert rubrics applied by a Gemini-2.5-Flash judge where rubric-scored, means over evaluation trials; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 4 | 79.57 |
| 2 | Qwen 3 235B A22B | 74.62 |
| 3 | DeepSeek V3.2 | 74.05 |
| 4 | Gemini 2.5 Pro | 69.42 |
| 5 | GPT-5.2 | 69.25 |
| 6 | Gemini 3 Pro (Preview) | 66.83 |
| 7 | GPT-4.1 | 64.58 |
| 8 | Kimi K2 (Thinking) | 63.67 |
| 9 | Qwen 3 8B | 62.39 |
| 10 | GLM-4.7 | 62.33 |
Interactive version: theaggregate.ai/benchmark?slug=dv-world-dv-inter-interaction-success · How It Works · Data refreshed daily, snapshot 2026-10-07.