Model profile
Ox Alpha
Evidence summary
Safety. Ox Alpha has an estimated Safety rank of #118; its 90% source-sensitivity interval is #15–#255. Published Safety evidence spans 5 eval lineages and 5 of 7 components. Its strongest relative result is SM-Bench (anti_hallucination, #1 of 84); its weakest is Olam Social Poker — Social Lie Rate (social_lie_rate_per_10000_turns, #17 of 22).
Freedom. Ox Alpha has an estimated Freedom rank of #465; its 90% source-sensitivity interval is #311–#531. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SM-Bench (overfit, #17 of 84); its weakest is SM-Bench (adversarial, #62 of 84).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Release source ↗curated outward evidence
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 |
|---|---|---|---|---|
| BullshitBench v2clear_pushback_rate | #20 / 114 | ↑0.585 | Source ↗official | |
| Manager Coercion Benchcoercion_ladder_depth | #8 / 35 | ↓6.767 | Source ↗self run | |
| Olam Social Poker — Social Lie Ratesocial_lie_rate_per_10000_turns | #17 / 22 | ↓30 | Source ↗official | |
| SM-Benchadversarial | #21 / 84 | ↑85.85 | Source ↗official | |
| SM-Benchambiguous_interpretation | #15 / 84 | ↑90.48 | Source ↗official | |
| SM-Benchanti_hallucination | #1 / 84 | ↑100 | Source ↗official | |
| SM-Bencheq_boundaries | #53 / 84 | ↑61.8 | Source ↗official | |
| SM-Benchoverfit | #17 / 84 | ↑88.52 | Source ↗official | |
| TACbase_welfare_rate | #63 / 84 | ↑21.15 | Source ↗self run |
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.