Gemini 3.1 Flash-Lite
Flash-Lite is Google's lowest-cost Gemini variant, tuned for high-volume, low-latency workloads like routing, classification and embedded agents.
Google DeepMind · Gemini
Gemini 3.1 Flash-Lite
Smallest, fastest, cheapest Gemini model
Flash-Lite is Google's lowest-cost Gemini variant, tuned for high-volume, low-latency workloads like routing, classification and embedded agents.
Key features
Gemini 3.1 Flash-Lite · AI Models
Context window
1M tokens
Max output
32K tokens
Released
March 3, 2026
Pricing
Lowest Gemini tier
Key features
- Lowest price-per-token in the Gemini family.
- Optimised for high-volume API workloads.
- Native multimodal input.
- Strong instruction-following despite reduced size.
Best for
Use Flash-Lite when you need Gemini quality at the lowest possible cost — moderation, intent classification, batch processing.
Frequently Asked Questions
What is Gemini 3.1 Flash-Lite best for?
Flash-Lite is ideal for moderation, intent classification, batch summarisation and any high-volume API workload where cost per query dominates.
Is Gemini 3.1 Flash-Lite still multimodal?
Yes — it accepts text, audio and image inputs natively, making it useful for lightweight multimodal classification and routing pipelines.
Open Chat
Flash-Lite is Google's lowest-cost Gemini variant, tuned for high-volume, low-latency workloads like routing, classification and embedded agents.