Model profile
GLM 4.7 Flash
Evidence summary
Safety. GLM 4.7 Flash has an estimated Safety rank of #174; its 90% source-sensitivity interval is #40–#317. Its behavior-only rank is #173; company governance moves the combined estimate to #174. Published Safety evidence spans 6 eval lineages and 3 of 7 components. Its strongest relative result is TAC (base_welfare_rate, #24 of 87); its weakest is AA-Omniscience (hallucination_rate, #319 of 345).
Freedom. GLM 4.7 Flash has an estimated Freedom rank of #21; its 90% source-sensitivity interval is #30–#114. Published Freedom evidence spans 3 eval lineages and 1 of 1 components. Its strongest relative result is UGI Leaderboard — base-model willingness (willingness_direct_score, #17 of 156); its weakest is UGI Leaderboard — base-model willingness (willingness_adherence_score, #51 of 156).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗glm-4-7-flash
- Hugging Face ↗zai-org/GLM-4.7-Flash
- OpenRouter ↗z-ai/glm-4.7-flash
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗repository created at proxy
Finetunes
Direct finetunes linked to this canonical base model. Quantized and repackaged derivatives are excluded.
huihui-ai/Huihui-GLM-4.7-Flash-abliterated ↗Direct non-quantized finetune documented by Hugging Face metadata
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 |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #319 / 345 | ↓0.9394 | Source ↗official | |
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #90 / 111 | ↑1413.0 | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #91 / 104 | ↓72.03 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #50 / 123 | ↑6.85 | Source ↗official | |
| SpeciEvalland_animal_4ns | #94 / 123 | ↓4.78 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #66 / 123 | ↓4.75 | Source ↗official | |
| SpeciEvalspeciesism | #113 / 123 | ↓2.85 | Source ↗official | |
| TACbase_welfare_rate | #24 / 87 | ↑33.33 | Source ↗self run | |
| Vectara HHEM Factual Consistencyfactual_consistency_rate | #41 / 94 | ↑90.7 | Source ↗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-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #14 / 104 | ↑72.03 | Source ↗official | |
| SpeechMap model completioncomplete_pct | #22 / 181 | ↑85.9 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_adherence_score | #51 / 156 | ↑4 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_direct_score | #17 / 156 | ↑5.5 | 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.