Model profile
K2 Chat
Evidence summary
Safety. K2 Chat 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, #110 of 272); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #199 of 272).
Freedom. K2 Chat has an estimated Freedom rank of #369; its 90% source-sensitivity interval is #188–#518. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #81 of 270); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #162 of 272).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Release source ↗repository created at proxy
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 | #199 / 272 | ↑12.66 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #141 / 272 | ↑87 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #110 / 272 | ↑81.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #190 / 270 | ↑94.41 | 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 | #128 / 272 | ↓87 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #162 / 272 | ↓81.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #81 / 270 | ↓94.41 | Source ↗official |