Model profile
Dbrx Instruct
Databricksdeveloper
2024-03-27release date
#302 / 312overall rank
7eval lineages
Evidence summary
Dbrx Instruct has an estimated overall rank of #302; its 90% source-sensitivity interval is #283–#305. Its behavior-only rank is #311; company governance moves the combined estimate to #302. Published evidence spans 7 evals and 5 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #163 of 241); its weakest is HELM Safety (harmbench, #80 of 80).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗databricks/dbrx-instruct
- OpenRouter ↗databricks/dbrx-instruct
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
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 | Source |
|---|---|---|---|---|
| AIRBench 2024 Safety Scenariossafety_scenarios | #80 / 80 | ↑0.254 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #187 / 241 | ↑11.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #163 / 241 | ↑85.17 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #218 / 241 | ↑34.44 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #192 / 239 | ↑91.23 | Source ↗official | |
| HELM Safetyanthropic_red_team | #80 / 80 | ↑0.766 | Source ↗official | |
| HELM Safetybbq | #67 / 80 | ↑0.792 | Source ↗official | |
| HELM Safetyharmbench | #80 / 80 | ↑0.271 | Source ↗official | |
| HELM Safetysimple_safety_tests | #80 / 80 | ↑0.535 | Source ↗official | |
| HELM Safetyxstest | #79 / 80 | ↑0.774 | Source ↗official |