SocialPersona: leaderboard
Personalized language-model assistants are often evaluated through a memory lens: can a model recall preferences users have explicitly stated in dialogue? More comprehensive personalization demands a harder capability -- inferring what users care about from the multimodal traces they naturally leave behind. We introduce SocialPersona, a benchmark for evaluating whether multimodal large language mo
Metric: Int. F1 (self-reported). Source: benchmarklist.com. Saturation forecast: Around May 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 41.44 | #79 |
| 2 | Gemini 2.5 Flash | 32.88 | #245 |
| 3 | GPT-4o Mini | 32.77 | #599 |
| 4 | Qwen 3 VL 8B Instruct | 32.19 | #420 |
| 5 | Qwen 3.5 35B A3B | 30.74 | #256 |
| 6 | Qwen 2.5 VL 7B Instruct | 17.72 | #654 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=socialpersona · How It Works · Data refreshed daily, snapshot 2026-10-20.