Model profile
Glm 4.6
Evidence summary
Glm 4.6 has an estimated overall rank of #123; its 90% source-sensitivity interval is #63–#223. Its behavior-only rank is #121; company governance moves the combined estimate to #123. Published evidence spans 5 evals and 6 of 7 behavior components. Its strongest relative result is SpeciEval (speciesism, #10 of 102); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #237 of 260).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗glm-4-6
- Hugging Face ↗zai-org/GLM-4.6
- OpenRouter ↗z-ai/glm-4.6
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #84 / 311 | 0.6613 | ↓ lower | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #189 / 260 | 12.66 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #237 / 260 | 74.67 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #176 / 260 | 55 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #155 / 258 | 95.45 | ↑ higher | Source ↗official | |
| LiveSecBenchethics | #27 / 43 | 36.13 | ↑ higher | Source ↗official | |
| LiveSecBenchfactuality | #14 / 43 | 58.59 | ↑ higher | Source ↗official | |
| LiveSecBenchlegality | #26 / 43 | 39.39 | ↑ higher | Source ↗official | |
| LiveSecBenchprivacy | #33 / 43 | 29.92 | ↑ higher | Source ↗official | |
| LiveSecBenchpsychological_health | #15 / 43 | 60.3 | ↑ higher | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #50 / 102 | 6.8 | ↑ higher | Source ↗official | |
| SpeciEvalland_animal_4ns | #65 / 102 | 4.67 | ↓ lower | Source ↗official | |
| SpeciEvalsea_animal_4ns | #73 / 102 | 4.92 | ↓ lower | Source ↗official | |
| SpeciEvalspeciesism | #10 / 102 | 1.43 | ↓ lower | Source ↗official | |
| TrustLLM contemporary collapsed applicationtrustllm | #3 / 8 | 0.62 | ↑ higher | Source ↗official |
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.