Model profile
Granite 3.0 8B Instruct
Evidence summary
Safety. Granite 3.0 8B Instruct does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 3 of 7 components. Its strongest relative result is BlueBench AttaQ-100 (attaq_harmlessness_reward_pct, #3 of 18); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #180 of 248).
Freedom. Granite 3.0 8B Instruct has an estimated Freedom rank of #447; its 90% source-sensitivity interval is #294–#539. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #104 of 248); its weakest is BlueBench AttaQ-100 (attaq_harmlessness_reward_pct, #16 of 18).
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 · ibm · first party
- 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 |
|---|---|---|---|---|
| BlueBench AttaQ-100attaq_harmlessness_reward_pct | #3 / 18 | ↑88.03 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #180 / 248 | ↑12.66 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #145 / 248 | ↑86.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #108 / 248 | ↑75 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #86 / 246 | ↑97.77 | 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 |
|---|---|---|---|---|
| BlueBench AttaQ-100attaq_harmlessness_reward_pct | #16 / 18 | ↓88.03 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #104 / 248 | ↓86.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #138 / 248 | ↓75 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #161 / 246 | ↓97.77 | Source ↗official |