Model profile
Granite 4.0 H Small
Evidence summary
Safety. Granite 4.0 H Small has an estimated Safety rank of #53; its 90% source-sensitivity interval is #9–#296. Its behavior-only rank is #48; company governance moves the combined estimate to #53. Published Safety evidence spans 3 eval lineages and 4 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #16 of 237); its weakest is AA-Omniscience (hallucination_rate, #238 of 338).
Freedom. Granite 4.0 H Small has an estimated Freedom rank of #569; its 90% source-sensitivity interval is #375–#589. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #128 of 239); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #219 of 237).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗granite-4-0-h-small
- Official model page ↗Family-level model document · ibm · first party
- Release source ↗curated outward evidence
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 |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #238 / 338 | ↓0.8785 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #93 / 239 | ↑18.6 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #109 / 239 | ↑88.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #64 / 239 | ↑85.56 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #16 / 237 | ↑99.64 | Source ↗official | |
| Vectara HHEM Factual Consistencyfactual_consistency_rate | #11 / 94 | ↑94.8 | 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 / 239 | ↓88.67 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #172 / 239 | ↓85.56 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #219 / 237 | ↓99.64 | Source ↗official |