Personal Superintelligence Index
An internal benchmark created by Meta to evaluate a model's superintelligence capabilities in scenarios of personal daily life.
Timeline (last 90 days)
Muse Wonder Full scores 59.1 on Meta's Personal Superintelligence Index, leading all models.
The API-available Muse Wonder scores 54.3 on Meta's Personal Superintelligence Index, behind Claude Opus 5.5's 55.2 and GPT-6 Astra's 54.5.
Meta created an internal benchmark, the Personal Superintelligence Index, to measure a model's usefulness for ordinary people's daily lives.
Vendors building their own benchmarks has become the norm in AI releases, but credibility is disputed, because the test sets, scoring criteria and data distributions are not public, making external reproduction impossible.
All Facts (4)
Meta created an internal benchmark, the Personal Superintelligence Index, to measure a model's usefulness for ordinary people's daily lives.
60%UnverifiedMuse Wonder Full scores 59.1 on Meta's Personal Superintelligence Index, leading all models.
50%UnverifiedThe API-available Muse Wonder scores 54.3 on Meta's Personal Superintelligence Index, behind Claude Opus 5.5's 55.2 and GPT-6 Astra's 54.5.
50%UnverifiedVendors building their own benchmarks has become the norm in AI releases, but credibility is disputed, because the test sets, scoring criteria and data distributions are not public, making external reproduction impossible.
50%