Google has introduced Gemini 3 Flash, the newest addition to its Gemini 3 model family. This release focuses on speed and efficiency while keeping advanced reasoning intact. It is now the default model in the Gemini app and AI Mode in Search, replacing Gemini 2.5 Flash.
Gemini 3 Flash delivers strong performance across benchmarks. It scored 90.4% on GPQA Diamond, a test of PhD-level reasoning, and 81.2% on MMMU Pro, which measures multimodal understanding.
It also uses 30% fewer tokens than Gemini 2.5 Pro, making it more efficient for everyday tasks. According to Google, the model is three times faster than 2.5 Pro at a fraction of the cost.
Pricing is set at USD 0.50 per million input tokens and $3 per million output tokens. Audio input remains at USD 1 per million tokens. This makes the new model a lower-cost option for developers and enterprises.
For developers, the model offers faster coding support and agentic workflows. On SWE-bench Verified, a benchmark for coding agents, Gemini 3 Flash achieved a 78% score, outperforming both Gemini 2.5 Pro and Gemini 3 Pro. It is also designed for multimodal tasks such as video analysis and visual Q&A.
Also Read: Google brings Gemini upgrades to Google Translate
Enterprises can access the model through Vertex AI and Gemini Enterprise. For everyday users, the rollout in the Gemini app and Search means quicker responses and more accurate results.
Gemini 3 Flash is already available globally. Stay tuned for more updates.
