Versions

1 version · Metric: Task success rate

Versions of GAIA2
VersionReleasedModelsResults
GAIA2 Public 800 tasksLatestFebruary 12, 202611

Recently reported results: GAIA2 Public 800 tasks

Latest version: GAIA2 Public 800 tasks

Recently reported results for GAIA2 Public 800 tasks
ModelScoreSource
Muse Glimmer 30B (high)43.3%dev.meta.ai for Muse Glimmer 30B (high) on GAIA2 Public 800 tasks (opens in a new tab)Task success rateReported Model pageMetaSelf-reportedRecord for Muse Glimmer 30B on GAIA2 Public 800 tasks