Google has delayed the release of its next-generation AI model, Gemini 3.5 Pro, until July, shifting from the previously communicated June launch window. Sources familiar with the matter report that Google is extending the development period to collect extensive feedback from early testers and refine the model. This adjustment follows remarks made at the May I/O developer conference, where CEO Sundar Pichai stated the model would launch "next month." A Google spokesperson declined to comment on the updated schedule. The postponement arrives as competition within the AI sector intensifies. Although Google's Gemini 3 iteration surpassed internal projections, rivals Anthropic and OpenAI have been advancing rapidly, particularly in coding performance, which is emerging as a critical enterprise use case. Google intends to counter these moves with Gemini 3.5 Pro, which is anticipated to excel in long-horizon tasks and agent-based workflows. Early testing is already underway on Google's Antigravity platform and the AI benchmarking site LMArena. Data from these environments is being used to optimize the model for practical applications. The new release will also integrate adjustments based on the recent Flash 3.5 model, specifically addressing user criticism regarding excessive token consumption. This confirms indications that Google is applying lessons from Flash to enhance the efficiency and architecture of the 3.5 Pro variant. The delay underscores the pressure on Google to deliver a highly competitive product that meets enterprise standards while maintaining operational efficiency.