Gemini 3.5 Flash (High): benchmark results

Gemini 3.5 Flash evaluated at the high reasoning-effort setting. Provider: Google. Released 2026-05-19. Access: API.

Unified ELO 1713 ± 1, rank #76 of 1761 rated models, from 156 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA APEX-Agents47.05Pass@1 (%)100
Computer Anthology Terminal Tasks (Gemini CLI)24.4pass@1 (%)100
AA-LCR81Accuracy (self-reported)98.3
UGI - Writing69.71Writing Score98.1
Chatbot Arena (Text - Math)1504Arena Score97.8
UGI - Natural Intelligence68.98NatInt Score97.4
UGI Leaderboard55.37UGI Score96.4
Chess Puzzles (Epoch AI)50Accuracy (%)96.3
AA Omniscience - Law57.6Accuracy (%)96
Chatbot Arena (Vision - Chinese)1351Arena Score95.6
Vals AI MedCode55.83Accuracy (%)95.5
AA IFBench76.33Accuracy (%)95.3

Interactive version: theaggregate.ai/model?slug=gemini-3-5-flash-high · How It Works · Data refreshed daily, snapshot 2026-09-05.