Model profile
Jamba 1 7 Large
Evidence summary
Jamba 1 7 Large has an estimated overall rank of #256; its 90% source-sensitivity interval is #224–#262. Its behavior-only rank is #260; company governance moves the combined estimate to #256. Published evidence spans 3 evals and 4 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #71 of 260); its weakest is AA-Omniscience (hallucination_rate, #291 of 311).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗jamba-1-7-large
- Official model page ↗Family-level model document · ai21 · first party
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #291 / 311 | 0.9375 | ↓ lower | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #97 / 105 | 0.05 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #71 / 260 | 23 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #238 / 260 | 73.17 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #204 / 260 | 46.11 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #202 / 258 | 91.82 | ↑ higher | 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.