Gemini 3.5 Flash Lite (Medium): benchmark results
Provider: Google. Released 2026-07-21. Access: API.
Aggregate Rating 1614 ± 20, rank #681 of 2201 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DelistBench (Closed-Book) | 66.5 | Overall joint accuracy within 7 days (%; all 1,200 security- | 75 |
| DelistBench - Announcement Status (Closed-Book) | 85.4 | Announcement-status accuracy (%; whether each of 1,200 secur | 75 |
| Wolfram LLM Benchmarking Project | 51.2 | Correct Functionality (%) | 64.8 |
| ParseBench - Visual Grounding | 46.59 | Element Pass Rate | 44.1 |
| Chinese Competitive Debating - Winner Prediction | 59.5 | Accuracy (%; the model reads a full match transcript and giv | 43.8 |
| Context Arena | 43.57 | Average Score (%) | 39.5 |
| DSV-Mem | 8.2 | Accuracy (%), 1,000 stateful visual-memory questions, mean o | 38.5 |
| ARC-AGI-2 | 5.28 | Accuracy (%) | 34.3 |
| ParseBench - Semantic Formatting | 48.06 | Semantic Formatting Score | 25.8 |
| ParseBench - Tables | 72.88 | GTRM Score (GriTS + TableRecordMatch) | 23.3 |
| ARC-AGI-1 | 32.33 | Accuracy (%) | 22.4 |
| ParseBench | 50.45 | Overall Score | 21.7 |
Interactive version: theaggregate.ai/model?slug=gemini-3-5-flash-lite-medium · How It Works · Data refreshed daily, snapshot 2026-11-12.