← Models

Model profile

Granite 4.0 Small

IBMdeveloper
2025-10-02release date
#160 / 267overall rank
6eval lineages

Evidence summary

Granite 4.0 Small has an estimated overall rank of #160; its 90% source-sensitivity interval is #59–#205. Its behavior-only rank is #161; company governance moves the combined estimate to #160. Published evidence spans 6 evals and 3 of 7 behavior components. Its strongest relative result is HELM Safety (anthropic_red_team, #13 of 80); its weakest is HELM Safety (xstest, #53 of 80).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

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-evalRankValueDistributionBetterSource
AIRBench 2024 Safety Scenariossafety_scenarios#34 / 800.716↑ higherSource ↗official
HELM Safetyanthropic_red_team#13 / 800.996↑ higherSource ↗official
HELM Safetybbq#46 / 800.925↑ higherSource ↗official
HELM Safetyharmbench#46 / 800.711↑ higherSource ↗official
HELM Safetysimple_safety_tests#48 / 800.98↑ higherSource ↗official
HELM Safetyxstest#53 / 800.946↑ higherSource ↗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.

UGI Political Values

DimensionValueDistribution
Political Lean-10.9
Government49.2
Diplomacy59
Economy45.7
Society57.8