Gemini 3 Flash: frontier intelligence built for speed
2 days ago
- #AI
- #Gemini 3 Flash
- Gemini 3 Flash is introduced as part of the Gemini 3 model family, offering frontier intelligence built for speed at a lower cost.
- Gemini 3 Flash combines Pro-grade reasoning with Flash-level latency, efficiency, and cost, making it ideal for everyday tasks and agentic workflows.
- The model is rolling out globally to developers via Google AI Studio, Gemini CLI, and Google Antigravity, and to everyone via the Gemini app and AI Mode in Search.
- Gemini 3 Flash demonstrates frontier performance on PhD-level reasoning benchmarks, outperforming Gemini 2.5 Pro and rivaling larger frontier models.
- It is highly efficient, using 30% fewer tokens on average than Gemini 2.5 Pro for everyday tasks while maintaining high performance.
- Gemini 3 Flash is priced at $0.50/1M input tokens and $3/1M output tokens, with audio input remaining at $1/1M input tokens.
- The model excels in iterative development, offering low latency and strong performance in coding, tool use, and multimodal capabilities.
- Companies like JetBrains, Bridgewater Associates, and Figma are already using Gemini 3 Flash to transform their businesses.
- Gemini 3 Flash is now the default model in the Gemini app, replacing 2.5 Flash, and is starting to roll out as the default model for AI Mode in Search.
- The model can analyze videos, images, and audio recordings, providing actionable plans, quizzes, and interactive experiences in near real-time.