As AI chatbots improve in accuracy and quality, the cost per input and output tokens can grow. To offer a solution for this, several AI companies have come out with cost-effective AI versions, such as Gemini Flash, which is optimized for speed and cost-effectiveness. Gemini Flash-Lite has even cheaper input and output tokens but lower reasoning capabilities.
Google recently announced Gemini 3.1 Flash-Lite, which is the newest version of the efficient model. Despite being the most cost-effective option, it now combines the reasoning and accuracy of Gemini 2.5 Flash with the speed and low costs of Gemini 2.5 Flash-Lite.
Gemini 3.1 Flash-Lite has landed.
— Google DeepMind (@GoogleDeepMind) March 3, 2026
It’s our most cost-efficient Gemini 3 series model yet, built for intelligence at scale. Here’s what’s new 🧵 pic.twitter.com/BzD2bdg3Dx
Comparing Gemini 3.1 Flash-Lite
When Google measured the output speed of Gemini 3.1 Flash-Lite, it achieved 363, which is much higher than Gemini 2.5 Flash at 249 and only slightly lower than Gemini 2.5 Flash-Lite at 366 (higher scores are better). For price, Gemini 3.1 Flash-Lite costs $0.25/1M input tokens and $1.50/1M output tokens, which is cheaper than Gemini 2.5 Flash ($0.30/1M input and $2.50/1M output) but more expensive than Gemini 2.5 Flash-Lite ($0.10/1M input and $0.40/1M output).
Even though the older version of Flash-Lite is cheaper and faster, 3.1 scores significantly higher in academic reasoning, scientific knowledge, and multimodal understanding than both Gemini 2.5 Flash and 2.5 Flash-Lite. So, it’s a fast and cost-effective model that also has strong reasoning skills.
GPT-5 Mini, Claude 4.5 Haiku, and Grok 4.1 Fast are similar models, but they’re all much slower in speed, and only Grok 4.1 Fast is more cost-effective. When it comes to reasoning and knowledge, Gemini 3.1 Flash-Lite outranks these models in most categories.
How to Use Flash-Lite
Gemini 3.1 Flash-Lite is currently available in AI Studio and Vertex AI. Users can control how much they want the AI to “think” for them, allowing them to choose when Flash-Lite steps in. High-volume, low-latency tasks work best for Flash-Lite because speed matters more than deep reasoning, and it costs less for those tasks.
AI Shapes Brand Marketing
People use AI for all types of tasks, including everything from quick translations to personalized recommendations. If AI platforms aren’t recognizing your brand, you’ll have a hard time reaching your audience. That’s where AI optimization comes in.
Contact Avenue Z today to experience how our strategies help brands rank in both AI responses and search engines.
We are the Agency for Influence
Discover new ways to drive revenue and build reputation for your brand.
