Gemini 1.5 Pro (0514): benchmark results

May 2024 Gemini 1.5 Pro API snapshot, tracked when sources report the dated model. Provider: Google. Released 2024-05-14. Access: API.

Unified ELO 1520 ± 1, rank #599 of 1392 rated models, from 18 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
RewardBench88.2Score (%)88
BenchBench82.95Aggregate Score (%)87.5
RewardBench Chat Hard80.59Accuracy (%)81.8
RewardBench Reasoning91.99Accuracy (%)77.5
RewardBench Safety87.91Accuracy (%)75.6
BigCodeBench43.8Pass@1 (%)71.4
SEAL - Spanish1069Score61.9
SEAL - Coding994Score40.7
SEAL - Math92.28Score40
RewardBench Chat92.32Accuracy (%)38.6
SEAL - Korean40.42Score16.7
SEAL - Agentic Tool Use (Enterprise)40.42Score15.2

Interactive version: theaggregate.ai/model?slug=gemini-1-5-pro-0514 · How It Works · Data refreshed daily, snapshot 2026-09-05.