Google has introduced Gemini 3.1 Flash-Lite, the fastest and most cost-efficient model in its Gemini 3 lineup. Designed for near-instant responses and improved reasoning, it aims to deliver strong performance while significantly undercutting competing models on price.
The details
Flash-Lite completes Google’s tiered Gemini 3 rollout, arriving weeks after the Pro model as a budget-focused option for high-volume tasks that do not require a flagship system.
The model posted a 12-point improvement on the Artificial Analysis Intelligence Index compared with its predecessor, outperforming even some larger previous-generation Gemini models on reasoning benchmarks.
Pricing is a major differentiator. Flash-Lite costs roughly one quarter of Anthropic’s Haiku and one eighth of Gemini 3.1 Pro, though output pricing is about three times higher than the Gemini 2.5 version it replaces.
Why important?
Low-cost, high-speed models are quickly becoming the primary competitive arena in AI. Flash-Lite’s benchmark results indicate Google can push prices down without dramatically sacrificing intelligence. Still, despite strong benchmark performance across the Gemini 3 family, it can be debated if Google’s consumer impact in 2026 has yet matched the outreach seen from Anthropic and OpenAI.
Sources: