Model profile
Evidence summary
Safety. Fugu Ultra v2 does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 3 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #24 of 270); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #235 of 270).
Freedom. Fugu Ultra v2 has an estimated Freedom rank of #297; its 90% source-sensitivity interval is #203–#382. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #36 of 270); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #241 of 270).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- OpenRouter ↗sakana/fugu-ultra-v2
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 |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #31 / 270 | ↑41.86 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #235 / 270 | ↑69 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #24 / 270 | ↑98.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #163 / 268 | ↑95.45 | 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 |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #36 / 270 | ↓69 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #241 / 270 | ↓98.33 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #101 / 268 | ↓95.45 | Source ↗official |