Model profile

Pixtral 12B

Mistral AIdeveloper

2024-09-11release date

Not rankedSafety rank

#248 / 645Freedom rank

Evidence summary

Safety. Pixtral 12B does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 1 of 7 components. Its strongest relative result is BlueBench AttaQ-100 (attaq_harmlessness_reward_pct, #16 of 18); its weakest is BlueBench AttaQ-100 (attaq_harmlessness_reward_pct, #16 of 18).

Freedom. Pixtral 12B has an estimated Freedom rank of #248; its 90% source-sensitivity interval is #32–#559. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is BlueBench AttaQ-100 (attaq_harmlessness_reward_pct, #3 of 18); its weakest is BlueBench AttaQ-100 (attaq_harmlessness_reward_pct, #3 of 18).

Compare this model

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

Official and reference links

OpenRouter ↗mistralai/pixtral-12b
Release source ↗repository created at proxy

Safety evals

Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.

Eval / sub-eval	Rank	Value	Distribution	Source
BlueBench AttaQ-100attaq_harmlessness_reward_pct	#16 / 18	↑82.45		Source ↗official

Freedom evals

Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.

Eval / sub-eval	Rank	Value	Distribution	Source
BlueBench AttaQ-100attaq_harmlessness_reward_pct	#3 / 18	↓82.45		Source ↗official