← Models

Model profile

Gemini 2 Flash Lite

2025-02-05release date
1eval lineages

Evidence summary

Published evidence spans 1 evals and 4 of 7 behavior components. Its strongest relative result is AILuminate General Purpose AI Chat (child_sexual_exploitation, #3 of 32); its weakest is AILuminate General Purpose AI Chat (non_violent_crimes, #19 of 32).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

Published eval results

Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.

Eval / sub-evalRankValueDistributionBetterSource
AILuminate General Purpose AI Chatchild_sexual_exploitation#3 / 323↓ lowerSource ↗official
AILuminate General Purpose AI Chatdefamation#13 / 3213↓ lowerSource ↗official
AILuminate General Purpose AI Chathate#15 / 328↓ lowerSource ↗official
AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne#9 / 3213↓ lowerSource ↗official
AILuminate General Purpose AI Chatintellectual_property#16 / 3212↓ lowerSource ↗official
AILuminate General Purpose AI Chatnon_violent_crimes#19 / 3219↓ lowerSource ↗official
AILuminate General Purpose AI Chatprivacy#14 / 3211↓ lowerSource ↗official
AILuminate General Purpose AI Chatsex_related_crimes#6 / 327↓ lowerSource ↗official
AILuminate General Purpose AI Chatsexual_content#9 / 3212↓ lowerSource ↗official
AILuminate General Purpose AI Chatspecialized_advice#9 / 3211↓ lowerSource ↗official
AILuminate General Purpose AI Chatsuicide_self_harm#13 / 3211↓ lowerSource ↗official
AILuminate General Purpose AI Chatviolent_crimes#9 / 3212↓ lowerSource ↗official