Gemini 3.1 Flash
Gemini 3.1 Flash brings the latest Gemini quality at high speed and low cost, with dedicated Audio, Image and Live variants for real-time products.
Google DeepMind · Gemini
Gemini 3.1 Flash
Fast Gemini for high-throughput multimodal apps
Gemini 3.1 Flash brings the latest Gemini quality at high speed and low cost, with dedicated Audio, Image and Live variants for real-time products.
Key features
Gemini 3.1 Flash · AI Models
Context window
1M tokens
Max output
64K tokens
Released
April 15, 2026 (Audio), Feb 26, 2026 (Image)
Pricing
Lower-cost Gemini tier
Key features
- Sub-second latency for chat and tool-using agents.
- Audio, Image and Live variants for real-time multimodal UX.
- 1M-token context window.
- Significantly cheaper than Gemini 3.1 Pro.
Best for
Pick Flash for chatbots, voice agents, image-aware UX and any product where fast multimodal responses matter more than absolute peak intelligence.
Frequently Asked Questions
Is Gemini 3.1 Flash multimodal?
Yes — Gemini 3.1 Flash natively handles text, audio, image and video with sub-second latency, plus dedicated Audio, Image and Live variants for real-time applications.
How does Gemini 3.1 Flash compare to Gemini 3.1 Pro?
Flash trades some reasoning depth for much faster responses and lower cost. Use Pro for the hardest tasks; use Flash for high-throughput products where speed matters more than peak intelligence.
Open Chat
Gemini 3.1 Flash brings the latest Gemini quality at high speed and low cost, with dedicated Audio, Image and Live variants for real-time products.