Model profile
Granite 4.0 H Micro
Evidence summary
Safety. Granite 4.0 H Micro 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 (toxicity_attack_non_success_rate, #36 of 246); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #151 of 248).
Freedom. Granite 4.0 H Micro has an estimated Freedom rank of #562; its 90% source-sensitivity interval is #386–#573. 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, #117 of 248); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #210 of 246).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗ibm-granite/granite-4.0-h-micro
- OpenRouter ↗ibm-granite/granite-4.0-h-micro
- Official model page ↗Family-level model document · ibm · first party
- Release source ↗repository created at proxy
Finetunes
Direct finetunes linked to this canonical base model. Quantized and repackaged derivatives are excluded.
huihui-ai/Huihui-granite-4.0-h-micro-abliterated ↗Direct non-quantized finetune documented by Hugging Face metadata
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 | #151 / 248 | ↑13.95 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #129 / 248 | ↑87.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #98 / 248 | ↑78.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #36 / 246 | ↑99.09 | 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 | #117 / 248 | ↓87.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #149 / 248 | ↓78.89 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #210 / 246 | ↓99.09 | Source ↗official |