AgentViSS - Interaction Outcome (Direct Vision): leaderboard
AgentViSS (2026) evaluates visual social intelligence: multimodal agents play roles in 240 multi-party scenarios and must read expressions, posture and emotion from images while pursuing role goals, with each task labeled achieved, partial or not achieved by three judge models. This board is the 0 to 100 score for achieving the social outcome of the role when the agent receives the images directly; higher is better.
Metric: Interaction-outcome task score (achieving the role's visually grounded social goal), the images given directly to the agent; normalized task score (0-100), each role-task instance labeled Achieved 2, Partially Achieved 0.5 or Not Achieved 0 by majority vote of three judges (Gemini 3.1 Pro Preview, GPT-5.4, Qwen3.5-27B), over 2,340 role-task instances in 240 multi-party social scenarios with group images and role portraits; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 88.8 | #161 |
| 2 | GPT-5.4 | 82.91 | #123 |
| 3 | Qwen 3.5 122B A10B | 77.56 | #331 |
| 4 | Qwen 3.5 27B | 76.15 | #359 |
| 5 | GLM-4.6V | 67.56 | #701 |
| 6 | Qwen 3.5 9B | 62.69 | #803 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=agentviss-interaction-outcome-direct-vision · How It Works · Data refreshed daily, snapshot 2026-11-13.