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

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

Unified ELO 1767 ± 55, rank #177 of 1841 rated models, from 35 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA MMLU-Pro89.54Accuracy (%)99.7
AA Omniscience - Software Engineering (SWE) - Swift84Accuracy (%)97.3
AA Omniscience - Law50.7Accuracy (%)96.8
AA LiveCodeBench85.71Pass@1 (%)96.5
AA Omniscience - Business43.3Accuracy (%)95.7
AA Omniscience - Humanities & Social Sciences49.6Accuracy (%)95.2
AA-Omniscience Accuracy47.07Accuracy (%)94.1
Epoch AI - Scicode49.88Score90.7
AA Omniscience - Software Engineering (SWE) - Java46Accuracy (%)90.3
AA GPQA Diamond88.69Accuracy (%)90
AA Omniscience - Software Engineering (SWE) - Go62Accuracy (%)89.9
AA Omniscience - Health37.8Accuracy (%)89.7

Interactive version: theaggregate.ai/model?slug=gemini-3-pro-preview-low · How It Works · Data refreshed daily, snapshot 2026-07-25.