Model profile
Gemini 2.5 Pro Exp
Googledeveloper
2025-03-25release date
#31 / 309overall rank
4eval lineages
Evidence summary
Gemini 2.5 Pro Exp has an estimated overall rank of #31; its 90% source-sensitivity interval is #4–#224. Its behavior-only rank is #30; company governance moves the combined estimate to #31. Published evidence spans 4 evals and 3 of 7 behavior components. Its strongest relative result is Confabulations (confabulation_rate, #3 of 52); its weakest is FORTRESS (average_risk_score, #39 of 49).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gemini-2-5-pro-03-25
- OpenRouter ↗google/gemini-2.5-pro-exp-03-25
- Official model documentation ↗Family-level model document · google · first party
- Release source ↗direct research preview identity date
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-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| CAIS Risk Indexhle_overconfidence | #37 / 50 | ↓71 | Source ↗official | |
| CAIS Risk Indexmask | #39 / 52 | ↓46.9 | Source ↗official | |
| Confabulationsconfabulation_rate | #3 / 52 | ↓3.96 | Source ↗official | |
| FORTRESSaverage_risk_score | #39 / 49 | ↓54.89 | Source ↗official | |
| FORTRESSover_refusal_score | #6 / 46 | ↓1.38 | Source ↗official | |
| MASKlying_probability_pct | #33 / 53 | ↓44.07 | Source ↗official |