Gemini 3.1 Pro (Preview) (High) — benchmark results

Gemini 3.1 Pro (Preview) evaluated at the high reasoning-effort setting. Provider: Google. Released 2026-02-19. Access: API.

Unified ELO 1986 ± 37, rank #27 of 1776 rated models, from 47 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
FutureEval19.84Unified Forecasting Score100
MedCode59.06Score (self-reported)100
SEAL - Humanity's Last Exam46.44Score100
UGI - Natural Intelligence74.94NatInt Score99.3
Chess Bench LLM2066Lichess Rating99
NonoBench63.3Overall Accuracy (%)98.9
LLM Chess (Saplin)1511.4ELO98.6
UGI - Writing68.22Writing Score98.1
Context Arena MRCR (2-needle)87.8AUC@1M (%)96.2
LisanBench0.45Mean Path Length / Current Maximum96
ATLAS68.52ATLAScore (self-reported)95.5
Context Arena MRCR (8-needle)47.5AUC@1M (%)95.5

Interactive version: theaggregate.ai/model?slug=gemini-3-1-pro-preview-high · How the rankings work · Data refreshed daily, snapshot 2026-07-22.