Gemini 2.5 Pro (03-25) — benchmark results

March 25 Gemini 2.5 Pro experimental snapshot. Provider: Google. Released 2025-03-25. Access: API.

Unified ELO 1808 ± 23, rank #135 of 1776 rated models, from 64 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
KOFFVQA89.67Overall Score (%)100
KORGym - Mathematical and Logical0.93Score100
KORGym - Puzzle0.94Score100
KORGym - Strategic0.84Score100
LLM2014 Logic 2025-0384.27Median Score100
OJBench38.91Score (self-reported)100
SEAL - VISTA54.65Score100
Visual-Language Understanding54.65Score (self-reported)100
KOFFVQA - Graph and Chart Understanding93.33Score (%)98.8
KOFFVQA - Document Understanding100Score (%)98.1
KOFFVQA - Table Understanding96.67Score (%)98.1
KOFFVQA - Object Attributes86Score (%)97.5

Interactive version: theaggregate.ai/model?slug=gemini-2-5-pro-03-25 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.