Model profile
Gemma 4 E4B
Evidence summary
Safety. Gemma 4 E4B has an estimated Safety rank of #158; its 90% source-sensitivity interval is #1–#317. Its behavior-only rank is #171; company governance moves the combined estimate to #158. Published Safety evidence spans 3 eval lineages and 4 of 7 components. Its strongest relative result is AA-Omniscience (hallucination_rate, #38 of 361); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #259 of 272).
Freedom. Gemma 4 E4B has an estimated Freedom rank of #368; its 90% source-sensitivity interval is #234–#446. 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, #32 of 272); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #195 of 270).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gemma-4-e4b
- Official model documentation ↗Family-level model document · google · 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 | #38 / 361 | ↓0.3094 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #259 / 272 | ↑6.2 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #240 / 272 | ↑66.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #94 / 272 | ↑85.56 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #74 / 270 | ↑98 | Source ↗official | |
| MT-JailBench CrescendoXsafety_score | #11 / 21 | ↑21.38 | 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 | #32 / 272 | ↓66.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #175 / 272 | ↓85.56 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #195 / 270 | ↓98 | Source ↗official | |
| MT-JailBench CrescendoXsafety_score | #11 / 21 | ↓21.38 | Source ↗official |