Model profile
Gemini 3.5 Flash Lite
Evidence summary
Safety. Gemini 3.5 Flash Lite has an estimated Safety rank of #128; its 90% source-sensitivity interval is #44–#234. Its behavior-only rank is #138; company governance moves the combined estimate to #128. Published Safety evidence spans 9 eval lineages and 5 of 7 components. Its strongest relative result is SM-Bench (adversarial, #6 of 92); its weakest is Pokee-Isaac model card — DTAP (combined_attack_success_rate, #6 of 6).
Freedom. Gemini 3.5 Flash Lite has an estimated Freedom rank of #102; its 90% source-sensitivity interval is #61–#349. Published Freedom evidence spans 3 eval lineages and 1 of 1 components. Its strongest relative result is SM-Bench (eq_boundaries, #7 of 92); its weakest is SM-Bench (adversarial, #87 of 92).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gemini-3-5-flash-lite
- OpenRouter ↗google/gemini-3.5-flash-lite
- 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 | #47 / 358 | ↓0.3436 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #19 / 122 | ↑0.63 | Source ↗official | |
| Inkling-Small model card — FORTRESSbenign_answer_rate | #5 / 10 | ↑95.5 | Source ↗official | |
| Inkling-Small model card — FORTRESSharmful_refusal_rate | #8 / 10 | ↑70.7 | Source ↗official | |
| Inkling-Small model card — StrongREJECTsafety_rate | #9 / 10 | ↑97.6 | Source ↗official | |
| Olam Social Poker — Social Lie Ratesocial_lie_rate_per_10000_turns | #15 / 24 | ↓13 | Source ↗official | |
| Opposite-Narrator Sycophancysycophancy_rate_pct | #6 / 24 | ↓1 | Source ↗official | |
| Pokee-Isaac model card — DTAPbenign_task_success_rate | #2 / 6 | ↑0.833 | Source ↗official | |
| Pokee-Isaac model card — DTAPcombined_attack_success_rate | #6 / 6 | ↓0.663 | Source ↗official | |
| SM-Benchadversarial | #6 / 92 | ↑89.76 | Source ↗official | |
| SM-Benchambiguous_interpretation | #60 / 92 | ↑82.44 | Source ↗official | |
| SM-Benchanti_hallucination | #70 / 92 | ↑86.91 | Source ↗official | |
| SM-Bencheq_boundaries | #7 / 92 | ↑76.12 | Source ↗official | |
| SM-Benchoverfit | #43 / 92 | ↑78.14 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #55 / 131 | ↑6.85 | Source ↗official | |
| SpeciEvalland_animal_4ns | #80 / 131 | ↓4.62 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #59 / 131 | ↓4.7 | Source ↗official | |
| SpeciEvalspeciesism | #110 / 131 | ↓2.58 | Source ↗official | |
| TACbase_welfare_rate | #63 / 92 | ↑23.72 | Source ↗self run |
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 |
|---|---|---|---|---|
| Inkling-Small model card — FORTRESSbenign_answer_rate | #5 / 10 | ↑95.5 | Source ↗official | |
| Inkling-Small model card — FORTRESSharmful_refusal_rate | #3 / 10 | ↓70.7 | Source ↗official | |
| Inkling-Small model card — StrongREJECTsafety_rate | #2 / 10 | ↓97.6 | Source ↗official | |
| SM-Benchadversarial | #87 / 92 | ↓89.76 | Source ↗official | |
| SM-Bencheq_boundaries | #7 / 92 | ↑76.12 | Source ↗official | |
| SM-Benchoverfit | #43 / 92 | ↑78.14 | Source ↗official | |
| SpeechMap model completioncomplete_pct | #24 / 181 | ↑84.9 | Source ↗official |