Model profile
Gemini 3.8 Flash
Evidence summary
Safety. Gemini 3.8 Flash has an estimated Safety rank of #25; its 90% source-sensitivity interval is #15–#88. Its behavior-only rank is #30; company governance moves the combined estimate to #25. Published Safety evidence spans 19 eval lineages and 7 of 7 components. Its strongest relative result is CAIS Risk Index (agent_red_teaming, #2 of 49); its weakest is Vals AI Cheating Audit (biomystery_bench_cheating_attempt_rate_pct, #9 of 9).
Freedom. Gemini 3.8 Flash has an estimated Freedom rank of #449; its 90% source-sensitivity interval is #252–#570. Published Freedom evidence spans 4 eval lineages and 1 of 1 components. Its strongest relative result is SM-Bench (overfit, #5 of 92); its weakest is SM-Bench (adversarial, #82 of 92).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gemini-3-8-flash-medium
- OpenRouter ↗google/gemini-3.8-flash
- Official model documentation ↗Family-level model document · google · first party
- Release source ↗direct research preview identity date
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 | #92 / 358 | ↓0.519 | Source ↗official | |
| BioSecBench-Refusal V2redteam_refusal_pct | #4 / 6 | ↑77.54 | Source ↗official | |
| BioSecBench-Refusal V2routine_compliance_pct | #4 / 6 | ↑36.61 | Source ↗official | |
| CAIS Risk Indexagent_red_teaming | #2 / 49 | ↓29.8 | Source ↗official | |
| CAIS Risk Indexbioweapons_assistance | #25 / 54 | ↓65.6 | Source ↗official | |
| CAIS Risk Indexhle_overconfidence | #25 / 55 | ↓51.3 | Source ↗official | |
| CAIS Risk Indexmachiavelli | #23 / 51 | ↓86.5 | Source ↗official | |
| CAIS Risk Indexmask | #39 / 57 | ↓41.4 | Source ↗official | |
| CAIS Risk Indexpolitical_manipulation | #25 / 51 | ↓46.3 | Source ↗official | |
| CAIS Risk Indextextquests_harm | #49 / 54 | ↓24.4 | Source ↗official | |
| CheatBench direct cheating propensitybiology_bioinformatics_cheating_rate_pct | #5 / 8 | ↓100 | Source ↗official | |
| CheatBench direct cheating propensityboard_games_cheating_rate_pct | #6 / 8 | ↓100 | Source ↗official | |
| CheatBench direct cheating propensitycreative_writing_cheating_rate_pct | #6 / 8 | ↓100 | Source ↗official | |
| CheatBench direct cheating propensityknowledge_work_cheating_rate_pct | #1 / 8 | ↓37 | Source ↗official | |
| CheatBench direct cheating propensitymathematical_research_cheating_rate_pct | #7 / 8 | ↓98 | Source ↗official | |
| CheatBench direct cheating propensitymenial_computation_cheating_rate_pct | #4 / 8 | ↓100 | Source ↗official | |
| CheatBench direct cheating propensitymultimodal_cheating_rate_pct | #7 / 8 | ↓100 | Source ↗official | |
| CheatBench direct cheating propensitysoftware_engineering_cheating_rate_pct | #1 / 8 | ↓5 | Source ↗official | |
| CheatBench direct cheating propensitysvg_competition_cheating_rate_pct | #7 / 8 | ↓100 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #52 / 270 | ↑32.3 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #246 / 270 | ↑63.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #44 / 270 | ↑96.11 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #127 / 268 | ↑96.55 | Source ↗official | |
| Google Gemini 3.8 launch — Gray Swan indirect prompt injection k=15attack_success_probability_k15_pct | #2 / 15 | ↓5.5 | Source ↗official | |
| Humanity's Last Exam RMS calibration error (Scale Labs)calibrationError | #12 / 42 | ↓51 | Source ↗official | |
| MentalHealthBenchoverall_model_performance_pct | #13 / 17 | ↑35.5 | Source ↗official | |
| Olam Diplomacy — Broken Promise Ratebroken_promise_rate_pct | #10 / 19 | ↓14.6 | Source ↗official | |
| Pander Scoreconversational_absolute_pander_score | #15 / 26 | ↓15.3 | Source ↗official | |
| Pander Scoreinstructional_absolute_pander_score | #20 / 26 | ↓66.04 | Source ↗official | |
| SM-Benchadversarial | #9 / 92 | ↑88.29 | Source ↗official | |
| SM-Benchambiguous_interpretation | #40 / 92 | ↑86.9 | Source ↗official | |
| SM-Benchanti_hallucination | #35 / 92 | ↑96.34 | Source ↗official | |
| SM-Bencheq_boundaries | #55 / 92 | ↑62.36 | Source ↗official | |
| SM-Benchoverfit | #5 / 92 | ↑95.08 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #58 / 131 | ↑6.83 | Source ↗official | |
| SpeciEvalland_animal_4ns | #36 / 131 | ↓4.33 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #65 / 131 | ↓4.72 | Source ↗official | |
| SpeciEvalspeciesism | #110 / 131 | ↓2.58 | Source ↗official | |
| TACbase_welfare_rate | #33 / 92 | ↑30.77 | Source ↗self run | |
| Vals AI Cheating Auditbiomystery_bench_cheating_attempt_rate_pct | #9 / 9 | ↓21.48 | Source ↗official | |
| Vals AI Cheating Auditswebench_verified_cheating_attempt_rate_pct | #2 / 6 | ↓11.6 | Source ↗official | |
| Vals AI Cheating Auditterminal_bench_cheating_shortcut_evidence_rate_pct | #12 / 14 | ↓2.622 | 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 |
|---|---|---|---|---|
| BioSecBench-Refusal V2redteam_refusal_pct | #3 / 6 | ↓77.54 | Source ↗official | |
| BioSecBench-Refusal V2routine_compliance_pct | #4 / 6 | ↑36.61 | Source ↗official | |
| CAIS Risk Indexbioweapons_assistance | #30 / 54 | ↑65.6 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #25 / 270 | ↓63.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #225 / 270 | ↓96.11 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #140 / 268 | ↓96.55 | Source ↗official | |
| SM-Benchadversarial | #82 / 92 | ↓88.29 | Source ↗official | |
| SM-Bencheq_boundaries | #55 / 92 | ↑62.36 | Source ↗official | |
| SM-Benchoverfit | #5 / 92 | ↑95.08 | Source ↗official |