EverMemBench (Full Context) - Fine-Grained Recall (Temporal): leaderboard
EverMemBench (Hu et al., 2026) tests long-horizon memory over simulated multi-party workplace chat: five year-long projects, about 845,000 tokens of group dialogue each, and 2,400 questions, here with the whole dialogue history in the model's context. Computing time spans between events from contexts with repeated phrases and adjacent dates; open-ended answers judged right or wrong by an LLM, accuracy in percent, higher is better.
Metric: Accuracy (%; whole multi-party dialogue history in context, open-ended, judged by an LLM). Source: arxiv.org. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 45 | #159 |
| 2 | GPT-4.1 Mini | 7 | #775 |
| 3 | Llama 4 Scout Instruct | 1.67 | #1532 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=evermembench-full-context-fine-grained-recall-temporal · How It Works · Data refreshed daily, snapshot 2026-11-12.