← Models

Model profile

Qwen3.7 Flash

Alibabadeveloper
2026-07-27release date
#97 / 348Safety rank
#655 / 663Freedom rank

Evidence summary

Safety. Qwen3.7 Flash has an estimated Safety rank of #97; its 90% source-sensitivity interval is #29–#281. Its behavior-only rank is #91; company governance moves the combined estimate to #97. Published Safety evidence spans 3 eval lineages and 2 of 7 components. Its strongest relative result is SpeciEval (speciesism, #36 of 133); its weakest is TAC (base_welfare_rate, #59 of 94).

Freedom. Qwen3.7 Flash has an estimated Freedom rank of #655; its 90% source-sensitivity interval is #488–#662. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SpeechMap model completion (complete_pct, #173 of 181); its weakest is SpeechMap model completion (complete_pct, #173 of 181).

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
Opposite-Narrator Sycophancysycophancy_rate_pct#10 / 24↓2.5Source ↗official
SpeciEvalbelief_animal_sentience#40 / 133↑6.92Source ↗official
SpeciEvalland_animal_4ns#46 / 133↓4.38Source ↗official
SpeciEvalsea_animal_4ns#47 / 133↓4.65Source ↗official
SpeciEvalspeciesism#36 / 133↓1.68Source ↗official
TACbase_welfare_rate#59 / 94↑24.36Source ↗self run

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
SpeechMap model completioncomplete_pct#173 / 181↑28.3Source ↗official