Gemini 3.5 Flash-Lite
model Your tags
Your notes
Google's high-volume, low-latency Gemini tier, with multimodal input, 1M context, built-in computer use, and throughput up to 350 output tokens/second. Google reports Terminal-Bench 2.1 at 54%, SWE-bench Pro 54.2%, OSWorld-Verified 74.0%, and GDM-MRCR v2 72.2%. Pricing starts at $0.30/M input and $2.50/M output tokens; parameter count is undisclosed.
Model Details
Context window 1,000,000
AA Intelligence 22
was 23 on v4.3
Benchmark Scores
| Benchmark | Score | Mode |
|---|---|---|
| Terminal-Bench 2.1 | 54% | — |
| SWE-bench Pro | 54.2% | — |
| OSWorld-Verified | 74.0% | — |