General
GoodAI LTM Benchmark
Released long-term-memory tests for conversational agents, preserving task definitions, scheduling conditions and complete recorded conversations.
452items
37subjects
The repository declares MIT. Constituent datasets retain their original attribution and terms; the upstream README also states usage restrictions for GPT-generated tests. Capturing these files does not establish additional permissions.license
long_term_memorydomain
textmodality
item-level responses released
Response matrix
Loading response matrix…
lowhighUnobserved
Scale: Item-specific grading scales