Gemini 2.5 Pro (Preview 06-05) — benchmark results

June 5 Gemini 2.5 Pro preview snapshot. Provider: Google. Released 2025-06-05. Access: API.

Unified ELO 1745 ± 40, rank #206 of 1776 rated models, from 43 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM Stats (Vibe-Eval)67.2Score (%)100
SEAL - VISTA54.63Score98.4
Visual-Language Understanding54.63Score (self-reported)98.1
SEAL - Fortress61.68Score96.4
IneqMath46Overall Accuracy (self-reported)96.3
Web-Bench25.3Pass@1 (%)95.7
SEAL - TutorBench55.65Score92.3
TutorBench55.65Score (self-reported)90.9
Wolfram LLM Benchmarking Project59.1Correct Functionality (%)88.9
Context Arena MRCR (2-needle)77.5AUC@1M (%)88.8
SpeechMap Compliance89% Requests Completed88.2
Fiction.LiveBench87.5Accuracy (%)87.5

Interactive version: theaggregate.ai/model?slug=gemini-2-5-pro-preview-06-05 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.