Model profile
AI21 Jamba Instruct v1.0
Evidence summary
Safety. AI21 Jamba Instruct v1.0 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 (cbrn_attack_non_success_rate, #107 of 270); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #245 of 270).
Freedom. AI21 Jamba Instruct v1.0 has an estimated Freedom rank of #153; its 90% source-sensitivity interval is #81–#435. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #44 of 270); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #164 of 270).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Official model page ↗Family-level model document · ai21 · first party
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 | #245 / 270 | ↑9.3 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #107 / 270 | ↑89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #224 / 270 | ↑43.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #187 / 268 | ↑94.45 | 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 | #164 / 270 | ↓89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #44 / 270 | ↓43.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #82 / 268 | ↓94.45 | Source ↗official |