Gemini 3.5 Pro: 2 Million Token Window & Deep Think AI

Google's Gemini 3.5 Pro is set to launch publicly later this month, following its preview in May. This new model is designed to replace Gemini Ultra as Google's most capable AI for complex reasoning and long tasks. Here's the thing: Gemini 3.5 Pro boasts a two-million token context window. This is the largest offered by any commercially available AI model, allowing users to submit the equivalent of several large books or thousands of research papers in one request. What's interesting is a feature called Deep Think. It applies step-by-step reasoning to complex tasks and was shown solving multi-step math problems and reviewing lengthy contract documents. This feature is aimed at professionals in legal, financial, and scientific fields. The model also offers expanded multimodal capabilities, processing text, images, audio, and video in a single request. This makes it useful for analyzing video footage with transcripts or reviewing engineering diagrams with technical specifications. The bottom line: this technology could significantly change how professionals handle large amounts of information and complex problems.