MemSyco-Bench - Objective Fact Judgment (Full Dialog): leaderboard

Metric: Generation accuracy (%; correct under the reference answer and task-specific rubric; objective factual questions whose historical user memory must not be used as evidence; 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

#ModelScore
1DeepSeek V4 Flash61.67
2Llama 3.3 70B Instruct57.33
3GPT-4o Mini54
4Llama 3.1 8B Instruct38.46
5Qwen 3 8B30.62

Interactive version: theaggregate.ai/benchmark?slug=memsyco-bench-objective-fact-judgment-full-dialog · How It Works · Data refreshed daily, snapshot 2026-09-29.