← Models

Model profile

Aurora Alpha

2026-02-09release date
Not rankedSafety rank
#587 / 645Freedom rank

Evidence summary

Safety. Aurora Alpha does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 3 of 7 components. Its strongest relative result is SM-Bench (ambiguous_interpretation, #72 of 84); its weakest is SM-Bench (adversarial, #80 of 84).

Freedom. Aurora Alpha has an estimated Freedom rank of #587; its 90% source-sensitivity interval is #320–#640. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is SM-Bench (adversarial, #4 of 84); its weakest is SM-Bench (overfit, #78 of 84).

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
SM-Benchadversarial#80 / 8470.73Source ↗official
SM-Benchambiguous_interpretation#72 / 8473.51Source ↗official
SM-Benchanti_hallucination#74 / 8479.06Source ↗official
SM-Bencheq_boundaries#77 / 8451.69Source ↗official
SM-Benchoverfit#78 / 8425.68Source ↗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
SM-Benchadversarial#4 / 8470.73Source ↗official
SM-Bencheq_boundaries#77 / 8451.69Source ↗official
SM-Benchoverfit#78 / 8425.68Source ↗official
SpeechMap model completioncomplete_pct#125 / 18145.6Source ↗official