Model profile
Seed Oss 36B Instruct
2025-08-20release date
2eval lineages
Evidence summary
Published evidence spans 2 evals and 4 of 7 behavior components. Its strongest relative result is LiveSecBench (ethics, #22 of 43); its weakest is LiveSecBench (legality, #36 of 43).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗seed-oss-36b-instruct
- Hugging Face ↗ByteDance-Seed/Seed-OSS-36B-Instruct
- NVIDIA ↗qc69jvmznzxy/seed-oss-36b-instruct
- OpenRouter ↗bytedance/seed-oss-36b-instruct
- 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 | #180 / 311 | 0.834 | ↓ lower | Source ↗official | |
| LiveSecBenchethics | #22 / 43 | 46.86 | ↑ higher | Source ↗official | |
| LiveSecBenchfactuality | #30 / 43 | 36.52 | ↑ higher | Source ↗official | |
| LiveSecBenchlegality | #36 / 43 | 19.65 | ↑ higher | Source ↗official | |
| LiveSecBenchprivacy | #36 / 43 | 21.14 | ↑ higher | Source ↗official | |
| LiveSecBenchpsychological_health | #34 / 43 | 27.32 | ↑ 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.