Skip to main content

Gemini 3.5 Flash-Lite

Google

๐Ÿง  Reasoning๐ŸŒ Web Search

Google's fastest, lowest-cost 3.5 model. Built for high-volume extraction, analysis, and autonomous subagent work.

Released July 21, 2026

Pricing

Input tokens$0.30/M
Output tokens$2.50/M

Capacity

Context window1.0M tokens
Max output66K tokens

Capabilities

โœ“Reasoning & Planning
โœ“Web Search
โœ—Open Source

Best Scores

GDPval-AA v21139.0 pts
SWE-bench Pro54.2%
Terminal-Bench 2.154.0%

Best For

๐Ÿ“summarization
๐Ÿ“Šanalysis
๐ŸŒtranslation

Benchmark Scores

Specialized Skills

GDPval-AA v2

Real paid work from 44 different jobs, such as law, nursing, and software. Judges compare two answers side by side without knowing which model wrote them, and the winner gains rating points. A typical human expert scores 1000, so a higher number means the work was picked over a human more often.

1139.0 pts

Independently measured by Artificial Analysis.

Software Engineering

SWE-bench Pro

The harder version of the coding test. Bigger codebases, trickier bugs. Scores drop for everyone, so the gaps between models become clearer.

54.2%

Provider-reported; no independent run recorded yet.

Terminal-Bench 2.1

Puts the AI in front of a computer terminal and asks it to finish multi-step tasks on its own. Measures how good an "AI agent" it is.

54.0%

Provider-reported; an independent run by Vals AI (Terminus 2) lands at 50.19%.

Why Choose Gemini 3.5 Flash-Lite?

Flash-Lite is designed for workloads where throughput and cost matter most. It keeps a 1M-token context window, reasoning, multimodal input, and built-in tools at a fraction of flagship pricing.

How It Compares

vs Gemini 3.6 Flash

Flash-Lite is five times cheaper on input and three times cheaper on output; 3.6 Flash is stronger for demanding agentic work.

vs Claude Haiku 4.5

Flash-Lite is cheaper with a much larger context window; Haiku posts a stronger published SWE-bench Verified score.

You Might Also Like