Amazon Nova Premier
Amazon's largest Nova model for complex tasks and model distillation
Amazon Nova Premier is the most capable model in Amazon's Nova family, built by Amazon for complex reasoning, planning and multi-step tasks. The company says it also serves as a teacher model for distilling smaller Nova models, and it is offered through Amazon Bedrock.
- Amazon's most capable Nova model, aimed at complex multi-step tasks
- Acts as a teacher model for distilling smaller Nova models
- Accepts text and images, and outputs text
- Available through Amazon Bedrock as a proprietary model
What it is
Amazon Nova Premier is a large language model from Amazon, positioned as the most capable member of the Nova family. Amazon built it for demanding work: complex reasoning, planning, and tasks that need several steps to complete. It accepts text and images and produces text, and it can work with tools and agents. A second role is model distillation: the company presents Premier as a teacher model, meaning its outputs can be used to train smaller, cheaper Nova models so they inherit some of its behaviour at lower serving cost. It is reached through Amazon Bedrock rather than as downloadable weights.
How it came about
Amazon developed the Nova line as its own family of foundation models for customers of its cloud platform. Premier arrived as the top tier of that family, above the smaller Nova models intended for speed and cost. Amazon announced it through its AWS News Blog, describing it both as the strongest Nova model for complex tasks and as a teacher for distillation into the rest of the range.
Inside the model
Amazon describes Nova Premier as a multimodal model: it reads text and images and writes text. It is built for multi-step reasoning and planning, and it can call tools as part of agentic workflows.
What it can do
Amazon positions Premier for complex, multi-step tasks that need planning and reasoning rather than simple single-turn answers. It handles both text and image inputs. The company also highlights its use in distillation, where it acts as a teacher for smaller Nova models.
How to use it
Nova Premier is offered through Amazon Bedrock, Amazon's managed service for running foundation models. It is a proprietary model: customers call it through the platform rather than downloading weights.
Why it matters
Within Amazon's own model family, Premier serves a structural role: it is the top model for hard tasks and the teacher used to distil capability into smaller, cheaper Nova models. That makes it both a product and a training source for the rest of the line.
Questions people ask about Amazon Nova Premier
What is Amazon Nova Premier?
It is the most capable model in Amazon's Nova family, built for complex reasoning, planning and multi-step tasks. Amazon also uses it as a teacher model, meaning smaller Nova models are trained from its outputs through a process called distillation.
Who makes it and how do I use it?
Amazon builds it and offers it through Amazon Bedrock, the company's managed platform for foundation models. It is proprietary, so you call it through the service rather than downloading model weights.
What can it take as input?
It is multimodal: it accepts text and images and produces text. Amazon positions it for complex tasks that involve several steps, including work where the model plans and calls tools.
What does 'teacher for model distillation' mean?
Distillation transfers behaviour from a large model to a smaller one. Amazon presents Premier as the teacher, so its outputs help train smaller Nova models that are cheaper and faster to run while keeping some of the larger model's capability.
Good reading and sources
There is no English Wikipedia article on Amazon Nova Premier yet, so the facts on this page come from the sources listed above: aws.amazon.com/blogs/aws/amazon-nova-premier-our-most-capable-model-fo. Some are the subject's own pages, so its claims about itself are reported as its claims. EveryLLM wrote the text in its own words, with AI assistance, and compares it with the sources to make sure no passage is copied.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.