MemSyco-Bench - Memory-Evidence Conflict (Full Dialog): leaderboard
Metric: Generation accuracy (%; correct under the reference answer and task-specific rubric; decisions where verified evidence conflicts with a remembered user preference; the whole preceding multi-turn dialogue is given in context with no memory system). Source: arxiv.org. Saturation forecast: Around December 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Flash | 59.67 |
| 2 | Llama 3.3 70B Instruct | 29 |
| 3 | GPT-4o Mini | 7 |
| 4 | Llama 3.1 8B Instruct | 4 |
| 5 | Qwen 3 8B | 0.67 |
Interactive version: theaggregate.ai/benchmark?slug=memsyco-bench-memory-evidence-conflict-full-dialog · How It Works · Data refreshed daily, snapshot 2026-09-29.