← Models

Model profile

Pangu Pro MoE 72B A16B

huaweideveloper
Not rankedSafety rank
#162 / 645Freedom rank

Evidence summary

Safety. Pangu Pro MoE 72B A16B does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 4 of 7 components. Its strongest relative result is LiveSecBench (psychological_health, #33 of 43); its weakest is LiveSecBench (privacy, #43 of 43).

Freedom. Pangu Pro MoE 72B A16B has an estimated Freedom rank of #162; its 90% source-sensitivity interval is #10–#562. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is LiveSecBench (privacy, #1 of 43); its weakest is LiveSecBench (psychological_health, #11 of 43).

Compare this model

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

Official and reference links

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-evalRankValueDistributionSource
LiveSecBenchethics#42 / 439.51Source ↗official
LiveSecBenchfactuality#42 / 4316.5Source ↗official
LiveSecBenchlegality#40 / 4312.05Source ↗official
LiveSecBenchprivacy#43 / 435.84Source ↗official
LiveSecBenchpsychological_health#33 / 4327.71Source ↗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-evalRankValueDistributionSource
LiveSecBenchethics#2 / 439.51Source ↗official
LiveSecBenchlegality#4 / 4312.05Source ↗official
LiveSecBenchprivacy#1 / 435.84Source ↗official
LiveSecBenchpsychological_health#11 / 4327.71Source ↗official