Gemini 3.5 Flash: Google Revolutionizes Ultra-Fast AI

Google Unveils Gemini 3.5: A Major Leap in Artificial Intelligence

Google has recently lifted the veil on Gemini 3.5, a new generation of artificial intelligence models that promises to combine cutting-edge intelligence with actionable capabilities. This launch marks a significant advancement in the development of intelligent agents capable of performing complex tasks. The flagship model of this series, 3.5 Flash, stands out for its exceptional performance, particularly in coding and long-term tasks, thus providing tangible utility in the real world.

3.5 Flash is now accessible to a wide audience around the globe. It is available to all users through the Gemini app and Google Search's AI mode. Developers can also access it via the Google Antigravity platform and the Gemini API in Google AI Studio and Android Studio. Additionally, businesses can integrate it within the Gemini Enterprise Agent Platform and Gemini Enterprise. Google also plans to launch 3.5 Pro next month, a model already in internal testing.

Unmatched Performance for Agents and Coding

The Gemini 3.5 Flash model is distinguished by an intelligence that rivals the best existing models while offering impressive execution speed. It outperforms the previous model, Gemini 3.1 Pro, on several demanding benchmarks such as Terminal-Bench 2.1 with a score of 76.2%, GDPval-AA with an Elo of 1656, and MCP Atlas with 83.6%. In terms of multimodal understanding, it achieves 84.2% on CharXiv Reasoning. In terms of speed, it generates tokens four times faster than other leading models.

Positioned in the upper right quadrant of the Artificial Analysis index, 3.5 Flash offers advanced intelligence at exceptional speed, proving that it is possible to reconcile quality and speed.

An Ideal Solution for Large-Scale Agentic Tasks

The balance between speed and performance makes 3.5 Flash a preferred tool for complex and long-term agentic tasks. What previously required days of work for a developer or weeks for an auditor can now be accomplished in a fraction of the time, often at a reduced cost compared to other leading models. Whether it's developing new applications, maintaining codebases, or preparing financial documents, 3.5 Flash plans, builds, and iterates rapidly.

With the updated Antigravity harness, 3.5 Flash becomes a powerful engine for deploying collaborative sub-agents capable of solving large-scale problems. Under supervision, it can reliably execute workflows and multi-step coding tasks while maintaining optimal performance.

Tangible Impact in the Professional World

The agentic capabilities of 3.5 Flash are already having a notable impact for developers and businesses. Google has closely collaborated with industry partners to identify challenges and complexities in their processes. The results are promising:

• Shopify uses sub-agents to analyze complex long-term data, improving the accuracy of growth forecasts for merchants worldwide.
• Macquarie Bank is exploring how 3.5 Flash can accelerate client onboarding by processing complex documents over 100 pages, enabling the retrieval of relevant information and formulating reliable recommendations with low latency.
• Salesforce integrates 3.5 Flash into Agentforce to automate complex business tasks, deploying multiple sub-agents that retain context and execute complex tool calls over multiple turns.

Personal AI Agents with 3.5 Flash

3.5 Flash is now the default model for the Gemini app and the AI mode in Search globally. At the I/O event, Google demonstrated how its agentic capabilities power new features to integrate advanced intelligence into daily life.

The new personal AI agent, Gemini Spark, utilizes 3.5 Flash to operate continuously, helping users navigate their digital lives and make decisions under their guidance. Gemini Spark is currently being rolled out to trusted testers, with a beta version expected for Google AI Ultra subscribers in the United States next week.

Security-Focused Development

Gemini 3.5 has been designed with Google's Frontier Safety Framework in mind. Cybersecurity and CBRN safety measures have been strengthened, reducing the risk of generating harmful content and errors in response to safe queries. This has been made possible through new advanced security training and interpretability tools that allow verification and understanding of the AI's internal reasoning before it provides a response.

Immediate Availability of 3.5 Flash

Gemini 3.5 Flash is now available via Google Antigravity, the Gemini API in Google AI Studio and Android Studio, as well as on the Gemini Enterprise Agent Platform and Gemini Enterprise. It is also accessible to everyone in the Gemini app and the AI mode in Search. The entire Gemini team is eager to see the innovations that users will create with this powerful tool.