Gemini 3 Pro (Preview) (Low): benchmark results

Provider: Google. Released 2025-11-18. Access: API.

Unified ELO 1629 ± 1, rank #343 of 1919 rated models, from 21 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA Omniscience - Law54.9Accuracy (%)94.6
AA Omniscience - Business42.7Accuracy (%)92.6
AA Omniscience - Humanities & Social Sciences49.8Accuracy (%)92.3
AA-Omniscience Accuracy48.32Accuracy (%)90.7
AA Omniscience - Health39.7Accuracy (%)87.3
AA GPQA Diamond88.69Accuracy (%)85.6
AA Omniscience - Software Engineering (SWE)59.6Accuracy (%)83.8
AA Omniscience1.8Score83.4
AA Omniscience - Science, Engineering & Mathematics43.2Accuracy (%)83.4
AA Humanity's Last Exam29.47Accuracy (%)81.4
AA Terminal-Bench Hard34.09Accuracy (%)78.8
AA Long Context Reasoning74Accuracy (%)75.5

Interactive version: theaggregate.ai/model?slug=gemini-3-pro-preview-low · How It Works · Data refreshed daily, snapshot 2026-09-08.