Gemini 3.5 Flash (Low): benchmark results
Provider: Google. Released 2026-05-19. Access: API.
Aggregate Rating 1707 ± 22, rank #341 of 2170 rated models, from 27 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Wolfram LLM Benchmarking Project | 67.7 | Correct Functionality (%) | 91.9 |
| ObviousBench | 99.31 | Answer pass³ (%) | 89.9 |
| Epoch AI - GPQA Diamond | 88.89 | Accuracy (%) | 81.5 |
| Chess Puzzles (Epoch AI) | 42 | Accuracy (%) | 77.2 |
| GRIPS-hard | 65.3 | Accuracy (%) | 73.1 |
| StudentBench - Learning Gain | 14.3 | GRE learning gain after one hour of AI tutoring (percentage | 72.7 |
| OTIS Mock AIME 2024-25 | 88.89 | Accuracy (%) | 71.7 |
| Epoch AI - Mystery Game Puzzles | 28 | Score | 71.3 |
| FinLifeBench - Financial State - Checkpoint State Accuracy | 75.4 | Cell-level state accuracy (%) | 70 |
| GRIPS | 91.6 | Accuracy (%) | 69.2 |
| MAG (Set-of-Mark) | 34.5 | Task success rate (%; functional WebArena checkers, LLM judg | 50 |
| MAG (Set-of-Mark) - Gated Guide Score | 20 | Gated guide score (%): zero on a failed task and 0.4 + 0.6 t | 50 |
Interactive version: theaggregate.ai/model?slug=gemini-3-5-flash-low · How It Works · Data refreshed daily, snapshot 2026-11-05.