← Benchmarks
MemoryBench
Internal recall evaluation over 64K-100K-token conversations containing distractors, conflicting facts, and later updates; LLM judge with weighted question-specific rubrics.
id memorybench · max 1 · 4 models reported
| # | Model | Score |
|---|---|---|
| 1 | Muse Wonder Meta | 0.86 |
| 2 | Muse Wonder Meta | 0.86 |
| 3 | Muse Wonder Full Meta | 0.85 |
| 4 | Muse Wonder Full Meta | 0.85 |