MuPPET - Utility: leaderboard
Metric: Utility (%): share of answers that use at least one useful preference or constraint extracted from the memories, judged by an LLM, in the main multi-party setting with no defence prompt: an LLM assistant acting for one user in a MuPPET multi-party workplace group conversation holds memory summaries with private information about that user and answers the group; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 82.03 |
| 2 | Qwen 3 14B | 81.14 |
| 3 | Qwen 3 8B | 78.29 |
| 4 | Gemini 2.5 Pro | 74.91 |
| 5 | Qwen 3 4B | 74.38 |
| 6 | Llama 3.1 8B Instruct | 69.93 |
| 7 | Llama 3 8B Instruct | 67.26 |
Interactive version: theaggregate.ai/benchmark?slug=muppet-utility · How It Works · Data refreshed daily, snapshot 2026-09-29.