Amazon Unveils Nova Premier: A Versatile AI Model for Complex Tasks and Model Distillation

Amazon has introduced Nova Premier, the latest and most advanced addition to its Nova family of AI models. Available through Amazon Bedrock, Nova Premier can process text, images, and videos, but not audio. Amazon highlights the model’s ability to handle complex tasks that require deep contextual understanding, multi-step planning, and precise execution across various tools and data sources. With a context window of 1 million tokens-equivalent to analyzing approximately 750,000 words at once-Nova Premier stands out for its scale. However, it lags behind top competitors like Google’s Gemini 2.5 Pro in coding, math, and science benchmarks, though it excels in knowledge retrieval and visual understanding according to Amazon’s internal tests. Pricing is competitive, matching Gemini 2.5 Pro for input tokens and slightly undercutting it for output tokens. Notably, Nova Premier is not designed as a ‘reasoning’ model, meaning it cannot take extra time to verify or fact-check its responses like some rival models. Amazon is positioning Nova Premier as a tool for distilling knowledge into smaller, more efficient models, making it especially useful for enterprises seeking to tailor AI to specific use cases. This launch underscores Amazon’s commitment to AI as a central pillar of its growth, with CEO Andy Jassy citing rapid expansion in generative AI applications and significant revenue growth in the sector.