Model profile
Seed OSS 36B Instruct
Evidence summary
Safety. Seed OSS 36B Instruct does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 4 of 7 components. Its strongest relative result is LiveSecBench (ethics, #22 of 43); its weakest is LiveSecBench (legality, #36 of 43).
Freedom. Seed OSS 36B Instruct has an estimated Freedom rank of #205; its 90% source-sensitivity interval is #58–#543. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is LiveSecBench (legality, #8 of 43); its weakest is UGI Leaderboard — base-model willingness (willingness_adherence_score, #143 of 156).
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
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 | #229 / 345 | ↓0.8639 | Source ↗official | |
| LiveSecBenchethics | #22 / 43 | ↑46.86 | Source ↗official | |
| LiveSecBenchfactuality | #30 / 43 | ↑36.52 | Source ↗official | |
| LiveSecBenchlegality | #36 / 43 | ↑19.65 | Source ↗official | |
| LiveSecBenchprivacy | #36 / 43 | ↑21.14 | Source ↗official | |
| LiveSecBenchpsychological_health | #34 / 43 | ↑27.32 | 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 |
|---|---|---|---|---|
| LiveSecBenchethics | #22 / 43 | ↓46.86 | Source ↗official | |
| LiveSecBenchlegality | #8 / 43 | ↓19.65 | Source ↗official | |
| LiveSecBenchprivacy | #8 / 43 | ↓21.14 | Source ↗official | |
| LiveSecBenchpsychological_health | #10 / 43 | ↓27.32 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_adherence_score | #143 / 156 | ↑0.6667 | Source ↗official | |
| UGI Leaderboard — base-model willingnesswillingness_direct_score | #44 / 156 | ↑4.333 | 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.