Model profile
ChatGLM3 Turbo
Z.aideveloper
2023-10-27release date
Not rankedSafety rank
Not rankedFreedom rank
Evidence summary
Safety. ChatGLM3 Turbo 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 CRiskEval (improvement_intent, #2 of 17); its weakest is CRiskEval (deception_willingness, #10 of 17).
Freedom. ChatGLM3 Turbo does not meet the evidence gate for a Freedom rank.
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Official model documentation ↗Family-level model document · zai · first party
- Release source ↗direct research preview identity date
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 |
|---|---|---|---|---|
| CRiskEvaldeception_willingness | #10 / 17 | ↓22.67 | Source ↗official | |
| CRiskEvaldesire_for_resource | #4 / 17 | ↓24.21 | Source ↗official | |
| CRiskEvalharmful_goal | #6 / 17 | ↓35.16 | Source ↗official | |
| CRiskEvalimprovement_intent | #2 / 17 | ↓43.25 | Source ↗official | |
| CRiskEvalmalicious_coordination | #7 / 17 | ↓10.08 | Source ↗official | |
| CRiskEvalself_preservation | #4 / 17 | ↓28.78 | Source ↗official | |
| CRiskEvalsituational_awareness | #4 / 17 | ↓54.39 | Source ↗official |
Freedom evals
No published sub-eval result contributes to this model’s Freedom profile.